* IOMMU setting
@ 2009-04-04 0:16 Eric Liu
2009-04-04 18:18 ` Muli Ben-Yehuda
0 siblings, 1 reply; 4+ messages in thread
From: Eric Liu @ 2009-04-04 0:16 UTC (permalink / raw)
To: kvm
Is there a quick way to check if system has IOMMU enabled in Linux?
I saw the following messages in /var/log/messages:
Apr 3 21:03:16 kernel: PCI-DMA: Disabling AGP.
Apr 3 21:03:16 kernel: PCI-DMA: aperture base @ f4000000 size 65536 KB
Apr 3 21:03:16 kernel: init_memory_mapping: 00000000f4000000-00000000f8000000
Apr 3 21:03:16 kernel: last_map_addr: f8000000 end: f8000000
Apr 3 21:03:16 kernel: PCI-DMA: using GART IOMMU.
Apr 3 21:03:16 kernel: PCI-DMA: Reserving 64MB of IOMMU area in the AGP aperture
Does this mean IOMMU is enabed? And i don't need anything like iommu=force in boot option, right?
CPU I used is the latest AMD Opteron Quad-core cpus. I didn't see any IOMMU setting option in BIOS, only enabling virtulization option is there. Is this the only option we need to enable?
Thanks
Eric
_________________________________________________________________
Rediscover Hotmail®: Get e-mail storage that grows with you.
http://windowslive.com/RediscoverHotmail?ocid=TXT_TAGLM_WL_HM_Rediscover_Storage1_042009
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: IOMMU setting
2009-04-04 0:16 IOMMU setting Eric Liu
@ 2009-04-04 18:18 ` Muli Ben-Yehuda
2009-04-05 17:14 ` Eric Liu
0 siblings, 1 reply; 4+ messages in thread
From: Muli Ben-Yehuda @ 2009-04-04 18:18 UTC (permalink / raw)
To: Eric Liu; +Cc: kvm
On Sat, Apr 04, 2009 at 12:16:50AM +0000, Eric Liu wrote:
>
> Is there a quick way to check if system has IOMMU enabled in Linux?
>
> I saw the following messages in /var/log/messages:
>
> Apr 3 21:03:16 kernel: PCI-DMA: Disabling AGP.
> Apr 3 21:03:16 kernel: PCI-DMA: aperture base @ f4000000 size 65536 KB
> Apr 3 21:03:16 kernel: init_memory_mapping: 00000000f4000000-00000000f8000000
> Apr 3 21:03:16 kernel: last_map_addr: f8000000 end: f8000000
> Apr 3 21:03:16 kernel: PCI-DMA: using GART IOMMU.
> Apr 3 21:03:16 kernel: PCI-DMA: Reserving 64MB of IOMMU area in the AGP aperture
>
> Does this mean IOMMU is enabed? And i don't need anything like
> iommu=force in boot option, right?
It means that you are running on an AMD system, and that this system
has a GART. You need an isolation-capable IOMMU such as Intel's VT-d
for KVM in-tree device passthrough.
Cheers,
Muli
--
Muli Ben-Yehuda | muli@il.ibm.com | +972-4-8281080
Manager, Virtualization and Systems Architecture
Master Inventor, IBM Haifa Research Laboratory
SYSTOR 2009---The Israeli Experimental Systems Conference
http://www.haifa.il.ibm.com/conferences/systor2009/
^ permalink raw reply [flat|nested] 4+ messages in thread
* RE: IOMMU setting
2009-04-04 18:18 ` Muli Ben-Yehuda
@ 2009-04-05 17:14 ` Eric Liu
2009-04-05 17:31 ` Muli Ben-Yehuda
0 siblings, 1 reply; 4+ messages in thread
From: Eric Liu @ 2009-04-05 17:14 UTC (permalink / raw)
To: muli; +Cc: kvm, joerg.roedel
Yes it is AMD system, however KVM website does mention AMD iommu support KVM device passthrough.
I want to make sure my system has this capability.
Regards
Eric
----------------------------------------
> Date: Sat, 4 Apr 2009 21:18:23 +0300
> From: muli@il.ibm.com
> To: ericliu2008@hotmail.com
> CC: kvm@vger.kernel.org
> Subject: Re: IOMMU setting
>
> On Sat, Apr 04, 2009 at 12:16:50AM +0000, Eric Liu wrote:
>>
>> Is there a quick way to check if system has IOMMU enabled in Linux?
>>
>> I saw the following messages in /var/log/messages:
>>
>> Apr 3 21:03:16 kernel: PCI-DMA: Disabling AGP.
>> Apr 3 21:03:16 kernel: PCI-DMA: aperture base @ f4000000 size 65536 KB
>> Apr 3 21:03:16 kernel: init_memory_mapping: 00000000f4000000-00000000f8000000
>> Apr 3 21:03:16 kernel: last_map_addr: f8000000 end: f8000000
>> Apr 3 21:03:16 kernel: PCI-DMA: using GART IOMMU.
>> Apr 3 21:03:16 kernel: PCI-DMA: Reserving 64MB of IOMMU area in the AGP aperture
>>
>> Does this mean IOMMU is enabed? And i don't need anything like
>> iommu=force in boot option, right?
>
> It means that you are running on an AMD system, and that this system
> has a GART. You need an isolation-capable IOMMU such as Intel's VT-d
> for KVM in-tree device passthrough.
>
> Cheers,
> Muli
> --
> Muli Ben-Yehuda | muli@il.ibm.com | +972-4-8281080
> Manager, Virtualization and Systems Architecture
> Master Inventor, IBM Haifa Research Laboratory
>
> SYSTOR 2009---The Israeli Experimental Systems Conference
> http://www.haifa.il.ibm.com/conferences/systor2009/
_________________________________________________________________
Rediscover Hotmail®: Get quick friend updates right in your inbox.
http://windowslive.com/RediscoverHotmail?ocid=TXT_TAGLM_WL_HM_Rediscover_Updates1_042009
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: IOMMU setting
2009-04-05 17:14 ` Eric Liu
@ 2009-04-05 17:31 ` Muli Ben-Yehuda
0 siblings, 0 replies; 4+ messages in thread
From: Muli Ben-Yehuda @ 2009-04-05 17:31 UTC (permalink / raw)
To: Eric Liu; +Cc: kvm, joerg.roedel
On Sun, Apr 05, 2009 at 05:14:44PM +0000, Eric Liu wrote:
>
> Yes it is AMD system, however KVM website does mention AMD iommu
> support KVM device passthrough. I want to make sure my system has
> this capability.
That's 'AMD IOMMU', as opposed to AMD GART, which is not an
isolation-capable IOMMU. I don't know if any chipsets with AMD's IOMMU
have been released yet.
Cheers,
Muli
--
Muli Ben-Yehuda | muli@il.ibm.com | +972-4-8281080
Manager, Virtualization and Systems Architecture
Master Inventor, IBM Haifa Research Laboratory
SYSTOR 2009---The Israeli Experimental Systems Conference
http://www.haifa.il.ibm.com/conferences/systor2009/
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2009-04-05 17:34 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-04-04 0:16 IOMMU setting Eric Liu
2009-04-04 18:18 ` Muli Ben-Yehuda
2009-04-05 17:14 ` Eric Liu
2009-04-05 17:31 ` Muli Ben-Yehuda
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).