Xen IOMMU disabled due to IVRS table... Blah blah blah

classic Classic list List threaded Threaded
25 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: Xen IOMMU disabled due to IVRS table... Blah blah blah

feral
FYI, I finally got a response from Asus Engineering.  Unfortunately they're all speaking something I am not, so we had to do 3 layers of translation both ways to get it done, but I believe the issue may now be resolved.
I sent them the ACPI dump and gave them the correct addresses for IOMMU 1 and 2 (or 0 and 1?).
I'm in the middle of moving into my new place, so haven't had a heap of time to test yet, but passthrough did work correctly (HDMI audio even works now) and no errors in dmesg.  I'm currently testing with KVM.

So that said, attached is the test BIOS we got hacked together.  Flash at your own risk.  I'm running it without any noticeable issues, but as I said, I haven't had time to do any rigorous testing or deep eval. on the actual IOMMU addresses (ie: haven't dumped ACPI/IVRS yet to confirm anything).




On Mon, Jul 15, 2013 at 4:14 PM, gizmochicken <[hidden email]> wrote:
I haven't yet tried it with vanilla Xen, but at least with XenServer 6.2 (which relies on Xen 4.1.5), IOMMU can be forced to initiate on my Asus M5A99FX Pro R2.0 motherboard (which suffers from an erroneous IVRS table) using the following:

      iommu=no-intremap

More information about this option, which seems to have been released for Intel setups, can be found here:  http://support.citrix.com/article/CTX136517







On Tue, May 28, 2013 at 12:49 PM, feral <[hidden email]> wrote:
> In my case, the handle id is wrong ( you can see what it should be on the
> support document http://support.amd.com/us/Processor_TechDocs/48882.pdf -
> per tables 77/79 it should be the ID of the IO-APIC). Unlike the conflict
> error, which can be worked around by passing another command line option,
> invalid IO-APIC entries in the IVRS table will always cause AMD-Vi to fail
> to be enabled if you have the XSA-36 patch installed.

This is the part I have an angry nerd rage issue with.  I'm pretty
sure I fall into the latter category where there is no option to
forcibly enable IOMMU.  I'd understand if the patch wasn't actually
removing functionality but in our case, it seems the patch does just
this even though the "bug" being addressed doesn't concern me (and
probably quite a few others) whatsoever.  The machine in question is
my gaming rig primarily, and gets a lot of use in QA, but nothing
mission critical.  I'd rather hobble along with a known bad IVRS table
if it has no affect on my work (or games) than be locked out entirely
:p.

I'm still a little confused though as I've heard a few people say that
this patch wasn't introduced until Xen 4.2 but I saw the issue first
pop up in Xen 4.1.2 and later.  Am I even looking at the right bug?
And what options are available to forcibly disable the checks other
than the "iommu=no-amd-iommu-perdev-intremap" ?




--
_____
Fact:
1. Ninjas are mammals.
2. Ninjas fight ALL the time.
3. The purpose of the ninja is to flip out and kill people.

_______________________________________________
Xen-users mailing list
[hidden email]
http://lists.xen.org/xen-users

SABERTOOTH-990FX-R20-ASUS-9901.CAP (10M) Download Attachment
SABERTOOTH-990FX-R20-ASUS-9901.ROM (10M) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Xen IOMMU disabled due to IVRS table... Blah blah blah

David Sutton-2
feral,

On Thu, Sep 5, 2013 at 2:05 PM, feral <[hidden email]> wrote:
FYI, I finally got a response from Asus Engineering.  Unfortunately they're all speaking something I am not, so we had to do 3 layers of translation both ways to get it done, but I believe the issue may now be resolved.
I sent them the ACPI dump and gave them the correct addresses for IOMMU 1 and 2 (or 0 and 1?).
I'm in the middle of moving into my new place, so haven't had a heap of time to test yet, but passthrough did work correctly (HDMI audio even works now) and no errors in dmesg.  I'm currently testing with KVM.

So that said, attached is the test BIOS we got hacked together.  Flash at your own risk.  I'm running it without any noticeable issues, but as I said, I haven't had time to do any rigorous testing or deep eval. on the actual IOMMU addresses (ie: haven't dumped ACPI/IVRS yet to confirm anything).

Thanks for getting this - unfortunately I have the R1.0 version of this board, so most likely won't be able to make use of it. Having said that, I'm watching progress of a patch which could help with this on the xen-devel list.

Regards,

  David


_______________________________________________
Xen-users mailing list
[hidden email]
http://lists.xen.org/xen-users
Reply | Threaded
Open this post in threaded view
|

Re: Xen IOMMU disabled due to IVRS table... Blah blah blah

gizmochicken
In reply to this post by feral
I also received a "test BIOS" from ASUS for my motherboard, which is
an ASUS M5A99FX PRO R2.0.  Like feral, I haven't done any rigorous
testing yet.  But from what I can tell, the "test BIOS" seems to have
cured all the IVRS errors that the board had been previously
experiencing.  Kudos to ASUS.

I'm also presently testing with KVM.  For those interested, here's a
link to where I describe passthrough of a Windows 7 guest to my Nvidia
GTX550Ti GPU using KVM/VFIO:
https://bbs.archlinux.org/viewtopic.php?pid=1313007#p1313007  (Note
that I haven't yet updated the description to reflect the fixed BIOS,
and so, at present, still refers to a kernel-based IVRS override that
I no longer use.)

The "test BIOS" is attached.  Flash at your own risk!  And certainly
don't flash to anything other than a ASUS M5A99FX PRO R2.0
motherboard.

Best regards,
GizmoChicken

On Thu, Sep 5, 2013 at 3:05 PM, feral <[hidden email]> wrote:

> FYI, I finally got a response from Asus Engineering.  Unfortunately they're
> all speaking something I am not, so we had to do 3 layers of translation
> both ways to get it done, but I believe the issue may now be resolved.
> I sent them the ACPI dump and gave them the correct addresses for IOMMU 1
> and 2 (or 0 and 1?).
> I'm in the middle of moving into my new place, so haven't had a heap of time
> to test yet, but passthrough did work correctly (HDMI audio even works now)
> and no errors in dmesg.  I'm currently testing with KVM.
>
> So that said, attached is the test BIOS we got hacked together.  Flash at
> your own risk.  I'm running it without any noticeable issues, but as I said,
> I haven't had time to do any rigorous testing or deep eval. on the actual
> IOMMU addresses (ie: haven't dumped ACPI/IVRS yet to confirm anything).
>
>
>
>
> On Mon, Jul 15, 2013 at 4:14 PM, gizmochicken <[hidden email]>
> wrote:
>>
>> I haven't yet tried it with vanilla Xen, but at least with XenServer 6.2
>> (which relies on Xen 4.1.5), IOMMU can be forced to initiate on my Asus
>> M5A99FX Pro R2.0 motherboard (which suffers from an erroneous IVRS table)
>> using the following:
>>
>>       iommu=no-intremap
>>
>> More information about this option, which seems to have been released for
>> Intel setups, can be found here:
>> http://support.citrix.com/article/CTX136517
>>
>>
>>
>>
>>
>>
>>
>> On Tue, May 28, 2013 at 12:49 PM, feral <[hidden email]> wrote:
>>>
>>> > In my case, the handle id is wrong ( you can see what it should be on
>>> > the
>>> > support document http://support.amd.com/us/Processor_TechDocs/48882.pdf
>>> > -
>>> > per tables 77/79 it should be the ID of the IO-APIC). Unlike the
>>> > conflict
>>> > error, which can be worked around by passing another command line
>>> > option,
>>> > invalid IO-APIC entries in the IVRS table will always cause AMD-Vi to
>>> > fail
>>> > to be enabled if you have the XSA-36 patch installed.
>>>
>>> This is the part I have an angry nerd rage issue with.  I'm pretty
>>> sure I fall into the latter category where there is no option to
>>> forcibly enable IOMMU.  I'd understand if the patch wasn't actually
>>> removing functionality but in our case, it seems the patch does just
>>> this even though the "bug" being addressed doesn't concern me (and
>>> probably quite a few others) whatsoever.  The machine in question is
>>> my gaming rig primarily, and gets a lot of use in QA, but nothing
>>> mission critical.  I'd rather hobble along with a known bad IVRS table
>>> if it has no affect on my work (or games) than be locked out entirely
>>> :p.
>>>
>>> I'm still a little confused though as I've heard a few people say that
>>> this patch wasn't introduced until Xen 4.2 but I saw the issue first
>>> pop up in Xen 4.1.2 and later.  Am I even looking at the right bug?
>>> And what options are available to forcibly disable the checks other
>>> than the "iommu=no-amd-iommu-perdev-intremap" ?
>>
>>
>
>
>
> --
> _____
> Fact:
> 1. Ninjas are mammals.
> 2. Ninjas fight ALL the time.
> 3. The purpose of the ninja is to flip out and kill people.

_______________________________________________
Xen-users mailing list
[hidden email]
http://lists.xen.org/xen-users

M5A99FX-PRO-R20-ASUS-9903.zip (4M) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Xen IOMMU disabled due to IVRS table... Blah blah blah

powerhouse64
I'm very interested to see whether or not Asus has fixed this BIOS bug. How is the test BIOS working?

Does anyone have an idea which boards are affected, or does it affect all Asus AMD boards?

How would I get this BIOS fix? To whom would I need to turn to at Asus?

For reference, I've written a little Xen VGA passthrough how-to on the Linux Mint forum and at least one user has encountered this exact issue - see  and go to the last post.
Reply | Threaded
Open this post in threaded view
|

Re: Xen IOMMU disabled due to IVRS table... Blah blah blah

feral
Sorry for the delay.
Attached is the test bios for the Sabertooth 990fx r2.0 (bios version 9901)

I've been using it successfully with KVM for most of a month at this point.


On Sat, Sep 14, 2013 at 3:22 AM, powerhouse64 <[hidden email]> wrote:
I'm very interested to see whether or not Asus has fixed this BIOS bug. How
is the test BIOS working?

Does anyone have an idea which boards are affected, or does it affect all
Asus AMD boards?

How would I get this BIOS fix? To whom would I need to turn to at Asus?

For reference, I've written a little Xen VGA passthrough how-to on the Linux
Mint forum and at least one user has encountered this exact issue - see
</a> <http://forums.linuxmint.com/viewtopic.php?f=42&t=112013>  and go to
the last post.



--
View this message in context: http://xen.1045712.n5.nabble.com/Xen-IOMMU-disabled-due-to-IVRS-table-Blah-blah-blah-tp5716461p5718631.html
Sent from the Xen - User mailing list archive at Nabble.com.

_______________________________________________
Xen-users mailing list
[hidden email]
http://lists.xen.org/xen-users



--
_____
Fact:
1. Ninjas are mammals.
2. Ninjas fight ALL the time.
3. The purpose of the ninja is to flip out and kill people.

_______________________________________________
Xen-users mailing list
[hidden email]
http://lists.xen.org/xen-users

9901.7z (4M) Download Attachment
12