From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Morton Subject: Re: [patch] fix boot with acpi=off Date: Mon, 10 Jul 2006 14:35:42 -0700 Message-ID: <20060710143542.e1be4877.akpm@osdl.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-path: Received: from smtp.osdl.org ([65.172.181.4]:5609 "EHLO smtp.osdl.org") by vger.kernel.org with ESMTP id S965245AbWGJVfq (ORCPT ); Mon, 10 Jul 2006 17:35:46 -0400 In-Reply-To: Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: "Brown, Len" Cc: vladimir.p.lebedev@intel.com, pavel@ucw.cz, linux-kernel@vger.kernel.org, linux-acpi@vger.kernel.org On Mon, 10 Jul 2006 14:37:53 -0400 "Brown, Len" wrote: > Dunno why acpi_lock_ac_dir() makes pavel's box fail w/ acpi=off. a) Because acpi_ac_init() forgot to check acpi_disabled, so acpi_ac_init() calls acpi_lock_ac_dir() even with acpi=off. b) Because acpi_lock_ac_dir() is doing down() on an uninitialised semaphore, which hangs. The fix to b) is in your tree now. The fix for a) is below. From: Pavel Machek With acpi=off and acpi_ac/battery compiled into kernel, acpi breaks boot. This fixes it. Signed-off-by: Pavel Machek Cc: "Brown, Len" Signed-off-by: Andrew Morton --- drivers/acpi/ac.c | 2 ++ drivers/acpi/battery.c | 3 +++ 2 files changed, 5 insertions(+) diff -puN drivers/acpi/ac.c~acpi-fix-boot-with-acpi=off drivers/acpi/ac.c --- a/drivers/acpi/ac.c~acpi-fix-boot-with-acpi=off +++ a/drivers/acpi/ac.c @@ -285,6 +285,8 @@ static int __init acpi_ac_init(void) { int result; + if (acpi_disabled) + return -ENODEV; acpi_ac_dir = acpi_lock_ac_dir(); if (!acpi_ac_dir) diff -puN drivers/acpi/battery.c~acpi-fix-boot-with-acpi=off drivers/acpi/battery.c --- a/drivers/acpi/battery.c~acpi-fix-boot-with-acpi=off +++ a/drivers/acpi/battery.c @@ -757,6 +757,9 @@ static int __init acpi_battery_init(void { int result; + if (acpi_disabled) + return -ENODEV; + acpi_battery_dir = acpi_lock_battery_dir(); if (!acpi_battery_dir) return -ENODEV; _