Hi Sean,
On 11/27/2017 03:25 PM, Mooney, Sean K wrote:
Sean can confirm, but my reading is that Sean's primary concern is with
the proposal to add any config option to enable iommu - not with the
proposal to change from a per port to per OVS basis.
[Mooney, Sean K] so I had a quick chat with ma
> > Conclusion: A global other_config parameter to enable/disable vhostuser
> > IOMMU is sufficient. By default this could be OFF for now and changed
> > to ON when broken Qemu versions are largely gone.
> [Mooney, Sean K] hi yes I just responded to this before I saw your reply.
> A global option
; Baudin ; Ilya Maximets ;
> Stokes, Ian ; Loftus, Ciara
> ; Darrell Ball ; Aaron Conole
>
> Subject: RE: [ovs-dev] [PATCH 2/2] netdev-dpdk: add support for vhost
> IOMMU feature
>
> > >> >> Hi Mark, All,
> > >> >>
> > >> >&g
; Baudin ; Ilya Maximets ;
> Stokes, Ian ; Loftus, Ciara
> ; Darrell Ball ; Aaron Conole
>
> Subject: Re: [ovs-dev] [PATCH 2/2] netdev-dpdk: add support for vhost
> IOMMU feature
>
> On 11/27/2017 12:55 PM, Kavanagh, Mark B wrote:
> >
> >
> >> From: Mooney, Sean
Ian
>; Loftus, Ciara ; Darrell Ball
>; Aaron Conole
>Subject: RE: [ovs-dev] [PATCH 2/2] netdev-dpdk: add support for vhost IOMMU
>feature
>
>> >> >> Hi Mark, All,
>> >> >>
>> >> >> I'm thinking about this and whether the current
> >> >> Hi Mark, All,
> >> >>
> >> >> I'm thinking about this and whether the current approach provides
> >> >> more than what is actually needed by users at the cost of making the
> >> >> user interface more complex.
> >[Mooney, Sean K] I am personally split on this. To enable iommu support in
> >
ner ; Franck
>> Baudin
>> ; Ilya Maximets ; Stokes, Ian
>> ; Loftus, Ciara ; Darrell Ball
>> ; Aaron Conole ; Mooney, Sean K
>>
>> Subject: RE: [ovs-dev] [PATCH 2/2] netdev-dpdk: add support for vhost IOMMU
>> feature
>>
>>
>>
>>>
>; Aaron Conole ; Mooney, Sean K
>
>Subject: RE: [ovs-dev] [PATCH 2/2] netdev-dpdk: add support for vhost IOMMU
>feature
>
>
>
>> -Original Message-
>> From: Kavanagh, Mark B
>> Sent: Friday, November 24, 2017 4:45 PM
>> To: Jan Scheurich ; Kevi
s, Ian
> ; Loftus, Ciara ; Darrell
> Ball ; Aaron Conole
> Subject: RE: [ovs-dev] [PATCH 2/2] netdev-dpdk: add support for vhost
> IOMMU feature
>
> Sounds good guys - I'll get cracking on this on Monday.
>
> Cheers,
> Mark
>
> >-Original Message-
&
e...@redhat.com; Flavio Leitner ; Franck Baudin
>; Mooney, Sean K ; Ilya Maximets
>; Stokes, Ian ; Loftus, Ciara
>; Darrell Ball ; Aaron Conole
>
>Subject: RE: [ovs-dev] [PATCH 2/2] netdev-dpdk: add support for vhost IOMMU
>feature
>
>+1
>Jan
>
>>
Maximets ; Ian Stokes
> ; Loftus, Ciara
> ; Darrell Ball ; Aaron Conole
> ; Jan Scheurich
>
> Subject: Re: [ovs-dev] [PATCH 2/2] netdev-dpdk: add support for vhost IOMMU
> feature
>
> On 11/16/2017 11:01 AM, Mark Kavanagh wrote:
> > DPDK v17.11 introduces support for the
On 11/16/2017 11:01 AM, Mark Kavanagh wrote:
> DPDK v17.11 introduces support for the vHost IOMMU feature.
> This is a security feature, that restricts the vhost memory
> that a virtio device may access.
>
> This feature also enables the vhost REPLY_ACK protocol, the
> implementation of which is k
On 11/24/2017 12:58 PM, Kavanagh, Mark B wrote:
>> From: Kevin Traynor [mailto:ktray...@redhat.com]
>> Sent: Friday, November 24, 2017 11:31 AM
>> To: Kavanagh, Mark B ; d...@openvswitch.org
>> Cc: maxime.coque...@redhat.com
>> Subject: Re: [ovs-dev] [PATCH 2/2] netd
On 11/24/2017 08:06 AM, Ilya Maximets wrote:
> Hello, Mark.
>
> Thanks for the patch. Few comments:
>
> 1. We can not change the RTE_VHOST_USER_IOMMU_SUPPORT flag if device already
>registered, i.e. if (dev->vhost_driver_flags & RTE_VHOST_USER_CLIENT) != 0.
>IMHO, we should print a warnin
>From: Kevin Traynor [mailto:ktray...@redhat.com]
>Sent: Friday, November 24, 2017 11:31 AM
>To: Kavanagh, Mark B ; d...@openvswitch.org
>Cc: maxime.coque...@redhat.com
>Subject: Re: [ovs-dev] [PATCH 2/2] netdev-dpdk: add support for vhost IOMMU
>feature
>
>On 11/16/2017
On 11/16/2017 11:01 AM, Mark Kavanagh wrote:
> DPDK v17.11 introduces support for the vHost IOMMU feature.
> This is a security feature, that restricts the vhost memory
> that a virtio device may access.
>
> This feature also enables the vhost REPLY_ACK protocol, the
> implementation of which is k
t; Subject: Re: [ovs-dev] [PATCH 2/2] netdev-dpdk: add support for vhost
> IOMMU feature
>
> > Hello, Mark.
> >
> > Thanks for the patch. Few comments:
> >
>
> Thanks for the feedback Ilya, great minds :). The feedback is quite
> timely, from my point of view I&
> Hello, Mark.
>
> Thanks for the patch. Few comments:
>
Thanks for the feedback Ilya, great minds :). The feedback is quite timely,
from my point of view I'd like to see these issues resolved as quickly as
possible. I don't think there's anything 'show stopper' wise in what you've
raise, we
Hello, Mark.
Thanks for the patch. Few comments:
1. We can not change the RTE_VHOST_USER_IOMMU_SUPPORT flag if device already
registered, i.e. if (dev->vhost_driver_flags & RTE_VHOST_USER_CLIENT) != 0.
IMHO, we should print a warning message in that case and should not update
the device
DPDK v17.11 introduces support for the vHost IOMMU feature.
This is a security feature, that restricts the vhost memory
that a virtio device may access.
This feature also enables the vhost REPLY_ACK protocol, the
implementation of which is known to work in newer versions of
QEMU (i.e. v2.10.0), bu
20 matches
Mail list logo
