linux-kernel - Re: [PATCH 1/2] dmar: fix fault interrupt setup

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <4CF5D990.7070407@jp.fujitsu.com>
Date:	Wed, 01 Dec 2010 14:13:52 +0900
From:	Kenji Kaneshige <kaneshige.kenji@...fujitsu.com>
To:	Suresh Siddha <suresh.b.siddha@...el.com>
CC:	"tglx@...utronix.de" <tglx@...utronix.de>,
	"mingo@...hat.com" <mingo@...hat.com>,
	"hpa@...or.com" <hpa@...or.com>, "x86@...nel.org" <x86@...nel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"indou.takao@...fujitsu.com" <indou.takao@...fujitsu.com>
Subject: Re: [PATCH 1/2] dmar: fix fault interrupt setup

(2010/12/01 5:18), Suresh Siddha wrote:
> On Tue, 2010-11-30 at 00:30 -0800, Kenji Kaneshige wrote:
>> Fix the problem dmar fault event is not notified in x2apic cluster mode.
>>
>> In the current code, dmar fault event is configured before setting up
>> the x86_cpu_to_logical_apicid percpu variable in x2apic cluster
>> mode. Because of this, invalid apic ID is used for dmar fault
>> interrupt and this cuases the problem.
>>
>> To fix the problem, do dmar fault event configuration after local apic
>> setup (after end_local_APIC_setup()).
>>
>> Signed-off-by: Kenji Kaneshige<kaneshige.kenji@...fujitsu.com>
>>
>>
>> ---
>>   arch/x86/kernel/apic/apic.c     |    9 +++++++++
>>   arch/x86/kernel/apic/probe_64.c |    7 -------
>>   2 files changed, 9 insertions(+), 7 deletions(-)
>>
>> Index: linux-next-20101125/arch/x86/kernel/apic/apic.c
>> ===================================================================
>> --- linux-next-20101125.orig/arch/x86/kernel/apic/apic.c
>> +++ linux-next-20101125/arch/x86/kernel/apic/apic.c
>> @@ -1745,6 +1745,15 @@ int __init APIC_init_uniprocessor(void)
>>
>>   	end_local_APIC_setup();
>>
>> +#ifdef CONFIG_INTR_REMAP
>> +	/*
>> +	 * Now that local APIC setup is completed, configure the fault
>> +	 * handling for interrupt remapping.
>> +	 */
>> +	if (intr_remapping_enabled)
>> +		enable_drhd_fault_handling();
>> +#endif
> 
> BTW, this is wrong. APIC_init_uniprocessor() gets called only for UP
> kernel. I will move it to the end_local_APIC_setup()
> 

Oh, it's totally wrong. I'm sorry about that.

Maybe even with my wrong patch, dmar interrupt is configured in the
intel_iommu_init()->init_dmars()->dmar_set_interrupt() code path. I
think this is why I could not noticed my badness in my test.

By the way, why do we need to try to configure dmar interrupt in the
intel_iommu_init() code path again?

Regards,
Kenji Kaneshige

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/