From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755478Ab0HCHU1 (ORCPT ); Tue, 3 Aug 2010 03:20:27 -0400 Received: from rcsinet10.oracle.com ([148.87.113.121]:28221 "EHLO rcsinet10.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755362Ab0HCHUZ (ORCPT ); Tue, 3 Aug 2010 03:20:25 -0400 Message-ID: <4C57C319.8070800@kernel.org> Date: Tue, 03 Aug 2010 00:19:53 -0700 From: Yinghai Lu User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.11) Gecko/20100714 SUSE/3.0.6 Thunderbird/3.0.6 MIME-Version: 1.0 To: "Eric W. Biederman" CC: Dave Airlie , LKML , Ingo Molnar Subject: Re: oops in ioapic_write_entry References: <4C577197.9020003@kernel.org> <4C57723C.1060400@kernel.org> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Source-IP: acsmt355.oracle.com [141.146.40.155] X-Auth-Type: Internal IP X-CT-RefId: str=0001.0A090204.4C57C32A.01C7,ss=1,fgs=0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 08/02/2010 08:13 PM, Eric W. Biederman wrote: > Yinghai Lu writes: > >> On 08/02/2010 06:32 PM, Yinghai Lu wrote: >>> On 08/02/2010 04:17 PM, Dave Airlie wrote: >>>>> >>>>> the kernel is using mptable, and the system have mcp55, so how come >>>>> with irq 35? >>>>> assume we should only have ioapic irq 0 - 23 ... >>>>> >>>>> Can you send out boot log with "debug apic=debug pci=routeirq" with >>>>> 2.6.32 and 2.6.35? >>>> >>>> Okay el6log is from a RHEL6 2.6.32 kernel, but it should give a good >>>> baseline, the 2.6.35 oops even earlier with all those options and is >>>> in the second attachment. >>> >> > > This patch is wrong and there is no reason to even suspect it will > affect this problem. At best this patch will trade one set of bugs > for another because at least on some platforms we always did something > like this. Having an irq 35 is odd and certainly a result of recent > changes, but in this case it doesn't look like it has anything to do > with the problem. > > Nacked-by: "Eric W. Biederman" > >> please use this one instead..., forget to run quilt refresh before sending it. >> >> [PATCH -v2] x86: fix pin_2_irq mapping >> >> We should not twist gsi to irq mapping if acpi is not used. >> >> -v2 remove not used irq_to_gsi() >> >> Signed-off-by: Yinghai Lu >> >> --- >> arch/x86/include/asm/io_apic.h | 10 ++++++++++ >> arch/x86/kernel/acpi/boot.c | 4 ++-- >> arch/x86/kernel/apic/io_apic.c | 5 +---- >> 3 files changed, 13 insertions(+), 6 deletions(-) >> >> Index: linux-2.6/arch/x86/include/asm/io_apic.h >> =================================================================== >> --- linux-2.6.orig/arch/x86/include/asm/io_apic.h >> +++ linux-2.6/arch/x86/include/asm/io_apic.h >> @@ -185,6 +185,16 @@ int mp_find_ioapic_pin(int ioapic, u32 g >> void __init mp_register_ioapic(int id, u32 address, u32 gsi_base); >> extern void __init pre_init_apic_IRQ0(void); >> >> +#ifdef CONFIG_ACPI >> +unsigned int gsi_to_irq(unsigned int gsi); >> +u32 irq_to_gsi(int irq); >> +#else >> +static inline unsigned int gsi_to_irq(unsigned int gsi) >> +{ >> + return gsi; >> +} >> +#endif >> + >> #else /* !CONFIG_X86_IO_APIC */ >> >> #define io_apic_assign_pci_irqs 0 >> Index: linux-2.6/arch/x86/kernel/acpi/boot.c >> =================================================================== >> --- linux-2.6.orig/arch/x86/kernel/acpi/boot.c >> +++ linux-2.6/arch/x86/kernel/acpi/boot.c >> @@ -100,7 +100,7 @@ static u32 isa_irq_to_gsi[NR_IRQS_LEGACY >> 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 >> }; >> >> -static unsigned int gsi_to_irq(unsigned int gsi) >> +unsigned int gsi_to_irq(unsigned int gsi) >> { >> unsigned int irq = gsi + NR_IRQS_LEGACY; >> unsigned int i; >> @@ -123,7 +123,7 @@ static unsigned int gsi_to_irq(unsigned >> return irq; >> } >> >> -static u32 irq_to_gsi(int irq) >> +u32 irq_to_gsi(int irq) >> { >> unsigned int gsi; >> >> Index: linux-2.6/arch/x86/kernel/apic/io_apic.c >> =================================================================== >> --- linux-2.6.orig/arch/x86/kernel/apic/io_apic.c >> +++ linux-2.6/arch/x86/kernel/apic/io_apic.c >> @@ -1029,10 +1029,7 @@ static int pin_2_irq(int idx, int apic, >> } else { >> u32 gsi = mp_gsi_routing[apic].gsi_base + pin; >> >> - if (gsi >= NR_IRQS_LEGACY) >> - irq = gsi; >> - else >> - irq = gsi_top + gsi; >> + irq = gsi_to_irq(gsi); >> } >> >> #ifdef CONFIG_X86_32 what is the point for making irq = gsi_top + gsi when mptable is used instead of acpi? Yinghai