在 2023/2/22 下午2:21, Michael S. Tsirkin 写道:
On Wed, Feb 22, 2023 at 10:34:39AM +0800, Heng Qi wrote:
The user will figure out how to mitigate when such QoS is not available. Either
to run in best-effort mode or mitigate differently.
Yes, our cloud security and cloud network team will
Hi, Jason. Long time no see. :)
在 2023/2/22 上午11:22, Jason Wang 写道:
在 2023/2/22 01:50, Michael S. Tsirkin 写道:
On Sat, Feb 18, 2023 at 10:37:15PM +0800, Heng Qi wrote:
+\subparagraph{Security risks between encapsulated packets and RSS}
+There may be potential security risks when encapsulated
On Wed, Feb 22, 2023 at 04:13:06AM +, Parav Pandit wrote:
>
> > From: Heng Qi
> > Sent: Tuesday, February 21, 2023 10:22 PM
>
> > +\item[VIRTIO_NET_F_VQ_NOTF_COAL(52)] Device supports virtqueue
> > notification coalescing.
> > +
> s/notification/notifications
> should be plural as multiple
On Wed, Feb 22, 2023 at 10:34:39AM +0800, Heng Qi wrote:
> > The user will figure out how to mitigate when such QoS is not available.
> > Either to run in best-effort mode or mitigate differently.
>
> Yes, our cloud security and cloud network team will configure and use inner
> hash on dpdk.
> From: Heng Qi
> Sent: Tuesday, February 21, 2023 10:22 PM
> +\item[VIRTIO_NET_F_VQ_NOTF_COAL(52)] Device supports virtqueue
> notification coalescing.
> +
s/notification/notifications
should be plural as multiple notifications are coalesced like the below
description of
在 2023/2/22 01:50, Michael S. Tsirkin 写道:
On Sat, Feb 18, 2023 at 10:37:15PM +0800, Heng Qi wrote:
+\subparagraph{Security risks between encapsulated packets and RSS}
+There may be potential security risks when encapsulated packets using RSS to
+select queues for placement. When a user inside
在 2023/2/22 上午7:18, Michael S. Tsirkin 写道:
On Tue, Feb 21, 2023 at 10:32:11PM +, Parav Pandit wrote:
From: Michael S. Tsirkin
Sent: Tuesday, February 21, 2023 4:46 PM
What is this information driver can't observe? It sees all the packets after
all,
we are not stripping tunneling
> From: Michael S. Tsirkin
> Sent: Tuesday, February 21, 2023 6:18 PM
> > The question of discussion was,
> > Scenario:
> > 1. device advertises the ability to hash on the inner packet header.
> > 2. device prefers that driver enable it only when it needs to use this extra
> packet parser in
On Tue, Feb 21, 2023 at 10:32:11PM +, Parav Pandit wrote:
>
> > From: Michael S. Tsirkin
> > Sent: Tuesday, February 21, 2023 4:46 PM
> >
> > What is this information driver can't observe? It sees all the packets
> > after all,
> > we are not stripping tunneling headers.
> Just the tunnel
> From: Michael S. Tsirkin
> Sent: Tuesday, February 21, 2023 4:46 PM
>
> What is this information driver can't observe? It sees all the packets after
> all,
> we are not stripping tunneling headers.
Just the tunnel type.
If/when that tunnel header is stripped, it gets complicated where
On Tue, Feb 21, 2023 at 09:36:06PM +, Parav Pandit wrote:
> > So you are saying either live with the problem (this is best effort yes?)
> Yes to best effort usage.
For sure something can be done to mitigate? How about randomizing the
key for example? That's in just like 1 minute of thinking.
> From: Michael S. Tsirkin
> Sent: Tuesday, February 21, 2023 4:24 PM
>
> On Tue, Feb 21, 2023 at 07:29:20PM +, Parav Pandit wrote:
> > > > When a specific receive queue is shared to receive packets of
> > > > multiple
> > > tunnels, there is no quality of service for packets of multiple
On Tue, Feb 21, 2023 at 07:29:20PM +, Parav Pandit wrote:
> > > When a specific receive queue is shared to receive packets of multiple
> > tunnels, there is no quality of service for packets of multiple tunnels.
> >
> > "shared to receive" is not grammatical either :)
> >
> "Shared by
> From: Michael S. Tsirkin
> Sent: Tuesday, February 21, 2023 12:06 PM
>
> On Tue, Feb 21, 2023 at 04:20:59AM +, Parav Pandit wrote:
> >
> > > From: Heng Qi
> > > Sent: Saturday, February 18, 2023 9:37 AM
> >
> > > If the tunnel is used to encapsulate the packets, the hash
> > >
On Tue, Feb 21, 2023 at 05:59:52PM +, Parav Pandit wrote:
>
> > From: Michael S. Tsirkin
> > Sent: Tuesday, February 21, 2023 12:52 PM
> >
> > On Tue, Feb 21, 2023 at 05:50:09PM +, Parav Pandit wrote:
> > >
> > >
> > > > From: Michael S. Tsirkin
> > > > Sent: Tuesday, February 21, 2023
> From: Michael S. Tsirkin
> Sent: Tuesday, February 21, 2023 12:52 PM
>
> On Tue, Feb 21, 2023 at 05:50:09PM +, Parav Pandit wrote:
> >
> >
> > > From: Michael S. Tsirkin
> > > Sent: Tuesday, February 21, 2023 12:42 PM
> > > >
> > > > What does "bits (for the driver)" mean? It made sense
> From: Michael S. Tsirkin
> Sent: Tuesday, February 21, 2023 12:44 PM
>
> On Tue, Feb 21, 2023 at 05:40:51PM +, Parav Pandit wrote:
> >
> > > From: Michael S. Tsirkin
> > > Sent: Tuesday, February 21, 2023 12:14 PM
> > > > The part that I am missing is, how do to reuse
> > > >
On Tue, Feb 21, 2023 at 05:50:09PM +, Parav Pandit wrote:
>
>
> > From: Michael S. Tsirkin
> > Sent: Tuesday, February 21, 2023 12:42 PM
> > >
> > > What does "bits (for the driver)" mean? It made sense together with
> > > "read-only", but I would drop "(for the driver)" as well.
> >
> >
On Sat, Feb 18, 2023 at 10:37:15PM +0800, Heng Qi wrote:
> +\subparagraph{Security risks between encapsulated packets and RSS}
> +There may be potential security risks when encapsulated packets using RSS to
> +select queues for placement. When a user inside a tunnel tries to control the
>
> From: Michael S. Tsirkin
> Sent: Tuesday, February 21, 2023 12:42 PM
> >
> > What does "bits (for the driver)" mean? It made sense together with
> > "read-only", but I would drop "(for the driver)" as well.
>
> Ouch Parav are you making search and replace changes without reading the
>
On Tue, Feb 21, 2023 at 05:40:51PM +, Parav Pandit wrote:
>
> > From: Michael S. Tsirkin
> > Sent: Tuesday, February 21, 2023 12:14 PM
> > > The part that I am missing is, how do to reuse virtio_net_hash_config and
> > > say
> > ignore all the existing fields related to rss, but only
On Tue, Feb 21, 2023 at 04:37:16PM +0100, Cornelia Huck wrote:
> On Fri, Feb 17 2023, Parav Pandit wrote:
>
> > Device configuration fields are read only. Avoid duplicating this
> > description for multiple fields.
> >
> > Instead describe it one time and do it in the driver requirements
> >
> From: Michael S. Tsirkin
> Sent: Tuesday, February 21, 2023 12:14 PM
> > The part that I am missing is, how do to reuse virtio_net_hash_config and
> > say
> ignore all the existing fields related to rss, but only consider
> hash_tunnel_types?
>
> Like a union? The answer is, don't. Just lay
> From: Heng Qi
> Sent: Tuesday, February 21, 2023 12:17 PM
>
> 在 2023/2/22 上午12:50, Parav Pandit 写道:
> >> From: Heng Qi
> >> Sent: Tuesday, February 21, 2023 11:44 AM
> >>> Patch-1 to introduce the feature bit, description, and link to CVQ
> dependency.
> >>> Patch-2 for its link in
On Tue, Feb 21, 2023 at 04:50:56PM +, Parav Pandit wrote:
>
> > From: Heng Qi
> > Sent: Tuesday, February 21, 2023 11:44 AM
>
> > > Patch-1 to introduce the feature bit, description, and link to CVQ
> > > dependency.
> > > Patch-2 for its link in virtio_net_config structure and
On Tue, Feb 21, 2023 at 04:20:59AM +, Parav Pandit wrote:
>
> > From: Heng Qi
> > Sent: Saturday, February 18, 2023 9:37 AM
>
> > If the tunnel is used to encapsulate the packets, the hash calculated using
> > the
> s/hash calculated/hash is calculated
>
> > outer header of the receive
> From: Heng Qi
> Sent: Tuesday, February 21, 2023 11:44 AM
> > Patch-1 to introduce the feature bit, description, and link to CVQ
> > dependency.
> > Patch-2 for its link in virtio_net_config structure and description.
> > Patch-3 for new command touching control VQ pieces.
>
> Yes, and you
On Fri, Feb 17 2023, Parav Pandit wrote:
> Device configuration fields are read only. Avoid duplicating this
> description for multiple fields.
>
> Instead describe it one time and do it in the driver requirements
> section.
>
> Fixes: https://github.com/oasis-tcs/virtio-spec/issues/161
>
> From: Heng Qi
> Sent: Tuesday, February 21, 2023 8:34 AM
> > Even without a queue overflow, this shared receive queue may not have a
> balanced number of packets.
> > For example, tunnel-2 occupied 90% of the queue and left only 10% for
> tunnel-1.
> > So, your example is right (and extreme),
在 2023/2/21 下午8:47, Parav Pandit 写道:
From: virtio-comm...@lists.oasis-open.org On Behalf Of Heng Qi
Also, a feature bit VIRTIO_NET_F_HASH_REPORT_TUNNEL are added to
report an encapsulation type, and the feature depends on
VIRTIO_NET_F_HASH_REPORT.
As we discussed that tunnel type alone is
在 2023/2/21 下午8:47, Parav Pandit 写道:
From: virtio-comm...@lists.oasis-open.org On Behalf Of Heng Qi
Also, a feature bit VIRTIO_NET_F_HASH_REPORT_TUNNEL are added to
report an encapsulation type, and the feature depends on
VIRTIO_NET_F_HASH_REPORT.
As we discussed that tunnel type alone is
> From: Parav Pandit
> Sent: Friday, February 17, 2023 10:45 AM
> Patch summary:
> patch-1: consolidate read only field at one place in driver requirements
> patch-2: define device configuration layout before describing its fields.
>
> changelog:
> v2->v3:
> - split into two patches
> - move
> From: virtio-comm...@lists.oasis-open.org open.org> On Behalf Of Heng Qi
> >> Also, a feature bit VIRTIO_NET_F_HASH_REPORT_TUNNEL are added to
> >> report an encapsulation type, and the feature depends on
> >> VIRTIO_NET_F_HASH_REPORT.
> > As we discussed that tunnel type alone is not
在 2023/2/21 下午7:48, David Edmondson 写道:
On Tuesday, 2023-02-21 at 16:38:52 +08, Heng Qi wrote:
...
+A device MAY set the coalescing parameter to a value close to a power of 2
value.
What is this about?
If it is intended to indicate that a device may use a value different to
that passed by
On Tuesday, 2023-02-21 at 16:38:52 +08, Heng Qi wrote:
> ...
> +A device MAY set the coalescing parameter to a value close to a power of 2
> value.
What is this about?
If it is intended to indicate that a device may use a value different to
that passed by the driver, more text to describe that
The current model of memory mapping at the back-end works fine with
Qemu, where a standard call to mmap() for the respective file
descriptor, passed from front-end, is generally all we need to do before
the front-end can start accessing the guest memory.
There are other complex cases though,
36 matches
Mail list logo