On Fri, Nov 15, 2019 at 12:27 PM Thomas Monjalon wrote:
>
> 07/11/2019 06:02, Jerin Jacob:
> > On Thu, Nov 7, 2019 at 4:03 AM Alex Williamson
> > wrote:
> > > Ideas have been suggested
> > > upstream for for quarantining VFs generated from user owned PFs such
> > > that we require an opt-in to ma
07/11/2019 06:02, Jerin Jacob:
> On Thu, Nov 7, 2019 at 4:03 AM Alex Williamson
> wrote:
> > Ideas have been suggested
> > upstream for for quarantining VFs generated from user owned PFs such
> > that we require an opt-in to make use of them in this way. Nobody
> > seems to be pursuing such ideas
On Thu, Nov 7, 2019 at 4:03 AM Alex Williamson
wrote:
>
> On Thu, 31 Oct 2019 18:03:53 +0100
> Thomas Monjalon wrote:
>
> > We don't get enough attention on this topic.
> > Let me rephrase the issue and the proposals with more people Cc'ed.
> >
> > We are talking about SR-IOV VFs in VMs
> > with
On Thu, 31 Oct 2019 18:03:53 +0100
Thomas Monjalon wrote:
> We don't get enough attention on this topic.
> Let me rephrase the issue and the proposals with more people Cc'ed.
>
> We are talking about SR-IOV VFs in VMs
> with a PF managed on the host by DPDK.
> The PF driver is either a (1) bifur
On Mon, 2019-11-04 at 11:16 +, Bruce Richardson wrote:
> On Fri, Nov 01, 2019 at 11:54:45AM +, Luca Boccassi wrote:
> > For distros, out-of-tree kernel modules are painful. From my POV,
> > it
> > would be preferable to try and find a solution upstream, even if it
> > is
> > going to be dif
On Fri, Nov 01, 2019 at 11:54:45AM +, Luca Boccassi wrote:
> For distros, out-of-tree kernel modules are painful. From my POV, it
> would be preferable to try and find a solution upstream, even if it is
> going to be difficult and require a lot of negotiation and work.
>
I don't think anyone
On Fri, Nov 1, 2019 at 5:24 PM Luca Boccassi wrote:
>
> For distros, out-of-tree kernel modules are painful. From my POV, it
I agree.
> would be preferable to try and find a solution upstream, even if it is
> going to be difficult and require a lot of negotiation and work.
I understand from RH,
For distros, out-of-tree kernel modules are painful. From my POV, it
would be preferable to try and find a solution upstream, even if it is
going to be difficult and require a lot of negotiation and work.
On Thu, 2019-10-31 at 18:03 +0100, Thomas Monjalon wrote:
> We don't get enough attention on
We don't get enough attention on this topic.
Let me rephrase the issue and the proposals with more people Cc'ed.
We are talking about SR-IOV VFs in VMs
with a PF managed on the host by DPDK.
The PF driver is either a (1) bifurcated (Mellanox case),
or (2) bound to UIO with igb_uio, or (3) bound to
> > >
> >
> > There is security issue in attaching DPDK PF driver and netdev bind to VF.
> > So this scheme is not upsteamble to Linux kernel. Since rte_flow had VF
> > action. We need this scheme to support VF action with VFIO. So, Out of tree
> > is the only way as it is DPDK specific feature. Al
On Wed, Oct 16, 2019 at 5:07 PM Jerin Jacob wrote:
>
> > >
> > >
> > > no-iommu case is different where we cannot screw Linux netdev driver, you
> > > can create a damage to your self that's an acceptable compromise.
> > >
> > > In this case, when DPDK PF bound application dies then it will impac
> >
> >
> > no-iommu case is different where we cannot screw Linux netdev driver, you
> > can create a damage to your self that's an acceptable compromise.
> >
> > In this case, when DPDK PF bound application dies then it will impact
> > netdev VF driver as gets stalled and there is a security is
On Thu, Oct 10, 2019 at 11:32 AM Jerin Jacob wrote:
>
>
>
> On Thu, 10 Oct, 2019, 4:58 AM Stephen Hemminger,
> wrote:
>>
>> On Tue, 8 Oct 2019 20:58:27 +0530
>> Jerin Jacob wrote:
>>
>> > On Tue, 8 Oct, 2019, 8:42 PM Stephen Hemminger,
>> >
>> > wrote:
>> >
>> > > On Fri, 6 Sep 2019 14:42:30
On Thu, 10 Oct, 2019, 4:58 AM Stephen Hemminger,
wrote:
> On Tue, 8 Oct 2019 20:58:27 +0530
> Jerin Jacob wrote:
>
> > On Tue, 8 Oct, 2019, 8:42 PM Stephen Hemminger, <
> step...@networkplumber.org>
> > wrote:
> >
> > > On Fri, 6 Sep 2019 14:42:30 +0530
> > > wrote:
> > >
> > > > From: Vamsi At
On Tue, 8 Oct 2019 20:58:27 +0530
Jerin Jacob wrote:
> On Tue, 8 Oct, 2019, 8:42 PM Stephen Hemminger,
> wrote:
>
> > On Fri, 6 Sep 2019 14:42:30 +0530
> > wrote:
> >
> > > From: Vamsi Attunuru
> > >
> > > The DPDK use case such as VF representer or OVS offload etc
> > > would call for PF a
On Tue, 8 Oct, 2019, 8:42 PM Stephen Hemminger,
wrote:
> On Fri, 6 Sep 2019 14:42:30 +0530
> wrote:
>
> > From: Vamsi Attunuru
> >
> > The DPDK use case such as VF representer or OVS offload etc
> > would call for PF and VF PCIe devices to bind vfio-pci
> > module to enable IOMMU protection.
>
On Fri, 6 Sep 2019 14:42:30 +0530
wrote:
> From: Vamsi Attunuru
>
> The DPDK use case such as VF representer or OVS offload etc
> would call for PF and VF PCIe devices to bind vfio-pci
> module to enable IOMMU protection.
>
> In addition to vSwitch use case, unlike, other PCI class of
> device
> -Original Message-
> From: Jerin Jacob Kollanukkaran
> Sent: Friday, September 6, 2019 6:58 PM
> To: Thomas Monjalon ; Vamsi Krishna Attunuru
>
> Cc: dev@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v1 1/1] kernel/linux: introduce vfio_pf kernel
> module
>
6:58 PM
To: Thomas Monjalon ; Vamsi Krishna Attunuru
Cc: dev@dpdk.org
Subject: RE: [dpdk-dev] [PATCH v1 1/1] kernel/linux: introduce vfio_pf kernel
module
-Original Message-
From: Thomas Monjalon
Sent: Friday, September 6, 2019 3:15 PM
To: Vamsi Krishna Attunuru
Cc: dev@dpdk.org
Attunuru
Cc: dev@dpdk.org
Subject: RE: [dpdk-dev] [PATCH v1 1/1] kernel/linux: introduce vfio_pf kernel
module
> -Original Message-
> From: Thomas Monjalon
> Sent: Friday, September 6, 2019 3:15 PM
> To: Vamsi Krishna Attunuru
> Cc: dev@dpdk.org; Jerin Jacob Kollanukkaran
> -Original Message-
> From: Thomas Monjalon
> Sent: Friday, September 6, 2019 3:15 PM
> To: Vamsi Krishna Attunuru
> Cc: dev@dpdk.org; Jerin Jacob Kollanukkaran
> Subject: Re: [dpdk-dev] [PATCH v1 1/1] kernel/linux: introduce vfio_pf kernel
> module
>
&g
06/09/2019 11:12, vattun...@marvell.com:
> From: Vamsi Attunuru
>
> The DPDK use case such as VF representer or OVS offload etc
> would call for PF and VF PCIe devices to bind vfio-pci
> module to enable IOMMU protection.
>
> In addition to vSwitch use case, unlike, other PCI class of
> devices,
22 matches
Mail list logo
