Hi Ben,
Just checking if you had a chance to look at v4 of this patch-set,
curious to know if you have any further comments.
Thanks,
-Harsha
On Thu, Jul 12, 2018 at 12:59 PM, Sriharsha Basavapatna
wrote:
> With the current OVS offload design, when an offload-device fails to add a
> flow rule
This is the first patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The patch detects OOR condition on a netdev port when ENOSPC error is
returned by TC-Flower while adding a flow rule. A new structure is added
to the netdev called "netdev_hw_info", to store OOR related
This is the second patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The packets-per-second (pps) rate for each flow is computed in the context
of revalidator threads when the flow stats are retrieved. The pps-rate is
computed only after a flow is revalidated and is not
This is the third patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The dynamic rebalancing functionality is implemented in this patch. The
ukeys that are not scheduled for deletion are obtained and passed as input
to the rebalancing routine. The rebalancing is done in the
With the current OVS offload design, when an offload-device fails to add a
flow rule and returns an error, OVS adds the rule to the kernel datapath.
The flow gets processed by the kernel datapath for the entire life of that
flow. This is fine when an error is returned by the device due to lack of
This is the fourth patch in the patch-set to support dynamic rebalancing
of offloaded flows.
A new OVS configuration parameter "offload-rebalance", is added to ovsdb.
The default value of this is "disable". To enable this feature, set the
value of this parameter to "pps-rate", which provides
On Tue, Jul 10, 2018 at 4:13 AM, Ben Pfaff wrote:
> On Sun, Jul 08, 2018 at 07:15:37PM +0530, Sriharsha Basavapatna via dev wrote:
>> This is the first patch in the patch-set to support dynamic rebalancing
>> of offloaded flows.
>>
>> The patch detects OOR condition o
With the current OVS offload design, when an offload-device fails to add a
flow rule and returns an error, OVS adds the rule to the kernel datapath.
The flow gets processed by the kernel datapath for the entire life of that
flow. This is fine when an error is returned by the device due to lack of
This is the first patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The patch detects OOR condition on a netdev port when ENOSPC error is
returned by TC-Flower while adding a flow rule. A new structure is added
to the netdev called "netdev_hw_info", to store OOR related
This is the third patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The dynamic rebalancing functionality is implemented in this patch. The
ukeys that are not scheduled for deletion are obtained and passed as input
to the rebalancing routine. The rebalancing is done in the
This is the second patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The packets-per-second (pps) rate for each flow is computed in the context
of revalidator threads when the flow stats are retrieved. The pps-rate is
computed only after a flow is revalidated and is not
This is the fourth patch in the patch-set to support dynamic rebalancing
of offloaded flows.
A new OVS configuration parameter "offload-rebalance", is added to ovsdb.
The default value of this is "disable". To enable this feature, set the
value of this parameter to "pps-rate", which provides
On Wed, Jul 11, 2018 at 3:33 AM, Ben Pfaff wrote:
> On Tue, Jul 10, 2018 at 11:59:35AM +0530, Sriharsha Basavapatna via dev wrote:
>> This is the first patch in the patch-set to support dynamic rebalancing
>> of offloaded flows.
>>
>> The patch detects OOR condition o
This is the third patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The dynamic rebalancing functionality is implemented in this patch. The
ukeys that are not scheduled for deletion are obtained and passed as input
to the rebalancing routine. The rebalancing is done in the
This is the first patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The patch detects OOR condition on a netdev port when ENOSPC error is
returned by TC-Flower while adding a flow rule. A new structure is added
to the netdev called "netdev_hw_info", to store OOR related
This is the second patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The packets-per-second (pps) rate for each flow is computed in the context
of revalidator threads when the flow stats are retrieved. The pps-rate is
computed only after a flow is revalidated and is not
v1 -> v2 changes:
- Fixed build errors reported by 0-day robot
- Updated patch prefixes with relevant subsystem names
**
With the current OVS offload design, when an offload-device fails to add a
flow rule and returns an error, OVS adds the rule to the kernel datapath.
The flow gets
This is the fourth patch in the patch-set to support dynamic rebalancing
of offloaded flows.
A new OVS configuration parameter "offload-rebalance", is added to ovsdb.
The default value of this is "disable". To enable this feature, set the
value of this parameter to "pps-rate", which provides
This is the second patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The packets-per-second (pps) rate for each flow is computed in the context
of revalidator threads when the flow stats are retrieved. The pps-rate is
computed only after a flow is revalidated and is not
This is the first patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The patch detects OOR condition on a netdev port when ENOSPC error is
returned by TC-Flower while adding a flow rule. A new structure is added
to the netdev called "netdev_hw_info", to store OOR related
With the current OVS offload design, when an offload-device fails to add a
flow rule and returns an error, OVS adds the rule to the kernel datapath.
The flow gets processed by the kernel datapath for the entire life of that
flow. This is fine when an error is returned by the device due to lack of
This is the third patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The dynamic rebalancing functionality is implemented in this patch. The
ukeys that are not scheduled for deletion are obtained and passed as input
to the rebalancing routine. The rebalancing is done in the
This is the fourth patch in the patch-set to support dynamic rebalancing
of offloaded flows.
A new OVS configuration parameter "offload-rebalance", is added to ovsdb.
The default value of this is "disable". To enable this feature, set the
value of this parameter to "pps-rate", which provides
Hi Kevin,
On Fri, Oct 12, 2018 at 10:08 PM Kevin Traynor wrote:
>
> On 09/27/2018 07:48 AM, Sriharsha Basavapatna via dev wrote:
> > This is the third patch in the patch-set to support dynamic rebalancing
> > of offloaded flows.
> >
> > The dynamic rebalancin
This is the first patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The patch detects OOR condition on a netdev port when ENOSPC error is
returned by TC-Flower while adding a flow rule. A new structure is added
to the netdev called "netdev_hw_info", to store OOR related
With the current OVS offload design, when an offload-device fails to add a
flow rule and returns an error, OVS adds the rule to the kernel datapath.
The flow gets processed by the kernel datapath for the entire life of that
flow. This is fine when an error is returned by the device due to lack of
This is the third patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The dynamic rebalancing functionality is implemented in this patch. The
ukeys that are not scheduled for deletion are obtained and passed as input
to the rebalancing routine. The rebalancing is done in the
This is the second patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The packets-per-second (pps) rate for each flow is computed in the context
of revalidator threads when the flow stats are retrieved. The pps-rate is
computed only after a flow is revalidated and is not
Hi Eelco,
Thanks for your review comments. Please see my response below. The
patch-set got merged earlier today. We will address your comments in
subsequent bug-fix patches.
On Fri, Oct 19, 2018 at 1:27 AM Eelco Chaudron wrote:
>
>
>
> On 18 Oct 2018, at 18:13, Sriharsha Basavap
A few lines with form feed characters (ASCII: ^L) were accidentally
deleted by a recent commit to support rebalancing of offloaded flows.
This patch reverts those lines.
Fixes: 57924fc91c ("revalidator: Rebalance offloaded flows")
Signed-off-by: Sriharsha Basavapatna
---
lib/dpif.c
On Wed, Oct 31, 2018 at 10:02 AM Ben Pfaff wrote:
>
> On Wed, Oct 31, 2018 at 09:01:59AM +0530, Sriharsha Basavapatna wrote:
> > On Wed, Oct 31, 2018 at 3:33 AM Ben Pfaff wrote:
> > >
> > > On Mon, Oct 29, 2018 at 10:16:45AM +0530, Sriharsha Basavapatna via dev
&g
A few lines with form feed characters (ASCII: ^L) were accidentally
deleted by a recent commit to support rebalancing of offloaded flows.
This patch reverts those lines.
Fixes: 57924fc91c ("revalidator: Rebalance offloaded flows")
Signed-off-by: Sriharsha Basavapatna
---
lib/dpif.c
On Thu, Nov 1, 2018 at 2:01 AM Ben Pfaff wrote:
>
> On Wed, Oct 31, 2018 at 01:11:58PM -0700, Ben Pfaff wrote:
> > On Thu, Nov 01, 2018 at 12:51:41AM +0530, Sriharsha Basavapatna via dev
> > wrote:
> > > A few lines with form feed characters (ASCII: ^L) were
patch
Thanks,
-Harsha
>
> On Thu, Nov 01, 2018 at 12:51:41AM +0530, Sriharsha Basavapatna via dev wrote:
> > A few lines with form feed characters (ASCII: ^L) were accidentally
> > deleted by a recent commit to support rebalancing of offloaded flows.
> > This patch r
On Wed, Oct 31, 2018 at 3:33 AM Ben Pfaff wrote:
>
> On Mon, Oct 29, 2018 at 10:16:45AM +0530, Sriharsha Basavapatna via dev wrote:
> > A few lines with form feed characters (ASCII: ^L) were accidentally
> > deleted by a recent commit to support rebalancing of offloaded flo
Hi Eelco,
Thanks for your review comments. Please see my response below.
On Fri, Oct 19, 2018 at 7:53 PM Eelco Chaudron wrote:
>
> On 18 Oct 2018, at 18:13, Sriharsha Basavapatna via dev wrote:
>
> This is the third patch in the patch-set to support dynamic rebalancing
> of
Hi Eelco,
Thanks for your comments, please see my response below.
On Fri, Oct 19, 2018 at 7:52 PM Eelco Chaudron wrote:
>
> On 18 Oct 2018, at 18:13, Sriharsha Basavapatna via dev wrote:
>
> > This is the first patch in the patch-set to support dynamic
> > rebalancing
&
Hi Eelco,
Thanks for your comments, please see my response below.
On Fri, Oct 19, 2018 at 7:52 PM Eelco Chaudron wrote:
>
> On 18 Oct 2018, at 18:13, Sriharsha Basavapatna via dev wrote:
>
> > This is the second patch in the patch-set to support dynamic
> > rebalancing
&
ating 1 entry, but we write to other offsets.
I've responded to your comments on the 3 patches. I've also answered
to your above concern on x2nrealloc(); there's no issue. Please see my
email on 3/3 for details.
Thanks,
-Harsha
>>
>> //Eelco
>>
>>
>> On 19 Oct 201
Hi Simon,
Thanks for your review comments. Please see my response inline.
On Mon, Sep 24, 2018 at 8:02 PM Simon Horman wrote:
>
> On Sat, Sep 15, 2018 at 09:40:08PM +0530, Sriharsha Basavapatna via dev wrote:
> > This is the first patch in the patch-set to support dynamic
Hi Simon,
Thanks for your review comments; please see my response inline.
On Mon, Sep 24, 2018 at 8:07 PM Simon Horman wrote:
>
> Hi Sriharsha,
>
> thanks for your patch.
>
> On Sat, Sep 15, 2018 at 09:40:09PM +0530, Sriharsha Basavapatna via dev wrote:
> > This is the s
. Please see comments inline.
>
> On Sat, Sep 15, 2018 at 09:40:10PM +0530, Sriharsha Basavapatna via dev wrote:
> > This is the third patch in the patch-set to support dynamic rebalancing
> > of offloaded flows.
> >
> > The dynamic rebalancing functi
With the current OVS offload design, when an offload-device fails to add a
flow rule and returns an error, OVS adds the rule to the kernel datapath.
The flow gets processed by the kernel datapath for the entire life of that
flow. This is fine when an error is returned by the device due to lack of
This is the third patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The dynamic rebalancing functionality is implemented in this patch. The
ukeys that are not scheduled for deletion are obtained and passed as input
to the rebalancing routine. The rebalancing is done in the
This is the first patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The patch detects OOR condition on a netdev port when ENOSPC error is
returned by TC-Flower while adding a flow rule. A new structure is added
to the netdev called "netdev_hw_info", to store OOR related
This is the second patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The packets-per-second (pps) rate for each flow is computed in the context
of revalidator threads when the flow stats are retrieved. The pps-rate is
computed only after a flow is revalidated and is not
With the current OVS offload design, when an offload-device fails to add a
flow rule and returns an error, OVS adds the rule to the kernel datapath.
The flow gets processed by the kernel datapath for the entire life of that
flow. This is fine when an error is returned by the device due to lack of
This is the third patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The dynamic rebalancing functionality is implemented in this patch. The
ukeys that are not scheduled for deletion are obtained and passed as input
to the rebalancing routine. The rebalancing is done in the
This is the first patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The patch detects OOR condition on a netdev port when ENOSPC error is
returned by TC-Flower while adding a flow rule. A new structure is added
to the netdev called "netdev_hw_info", to store OOR related
This is the second patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The packets-per-second (pps) rate for each flow is computed in the context
of revalidator threads when the flow stats are retrieved. The pps-rate is
computed only after a flow is revalidated and is not
On Sat, Sep 1, 2018 at 1:05 AM, Ben Pfaff wrote:
> On Thu, Jul 12, 2018 at 12:59:34PM +0530, Sriharsha Basavapatna via dev wrote:
>> This is the fourth patch in the patch-set to support dynamic rebalancing
>> of offloaded flows.
>>
>> A new OVS configuration p
On Sat, Sep 1, 2018 at 12:58 AM, Ben Pfaff wrote:
> On Thu, Jul 12, 2018 at 12:59:33PM +0530, Sriharsha Basavapatna via dev wrote:
>> This is the third patch in the patch-set to support dynamic rebalancing
>> of offloaded flows.
>>
>> The dynamic rebalancing f
This is the second patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The packets-per-second (pps) rate for each flow is computed in the context
of revalidator threads when the flow stats are retrieved. The pps-rate is
computed only after a flow is revalidated and is not
This is the first patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The patch detects OOR condition on a netdev port when ENOSPC error is
returned by TC-Flower while adding a flow rule. A new structure is added
to the netdev called "netdev_hw_info", to store OOR related
On Fri, Aug 31, 2018 at 10:04 PM, Ben Pfaff wrote:
> On Thu, Jul 12, 2018 at 12:59:31PM +0530, Sriharsha Basavapatna via dev wrote:
>> This is the first patch in the patch-set to support dynamic rebalancing
>> of offloaded flows.
>>
>> The patch detects OOR condition o
On Fri, Aug 31, 2018 at 11:22 PM, Ben Pfaff wrote:
> On Thu, Jul 12, 2018 at 12:59:32PM +0530, Sriharsha Basavapatna via dev wrote:
>> This is the second patch in the patch-set to support dynamic rebalancing
>> of offloaded flows.
>>
>> The packets-per-second (pps) rat
With the current OVS offload design, when an offload-device fails to add a
flow rule and returns an error, OVS adds the rule to the kernel datapath.
The flow gets processed by the kernel datapath for the entire life of that
flow. This is fine when an error is returned by the device due to lack of
This is the third patch in the patch-set to support dynamic rebalancing
of offloaded flows.
The dynamic rebalancing functionality is implemented in this patch. The
ukeys that are not scheduled for deletion are obtained and passed as input
to the rebalancing routine. The rebalancing is done in the
Hi Eli,
Thanks for this full-offload patch set. I have some comments inline.
Thanks,
-Harsha
On Wed, Nov 20, 2019 at 9:07 PM Eli Britstein wrote:
>
> Refactor the flow patterns and actions code to a new source files for
> better readability and towards adding more code to it.
>
>
On Wed, Nov 20, 2019 at 9:07 PM Eli Britstein wrote:
>
> Instead of statically allocated pattern items on the stack,dynamically
> allocate only the required items while parsing the required matches.
What are the benefits of dynamic allocation scheme? It is not clear
from the patch whether this
On Wed, Nov 20, 2019 at 9:05 PM Eli Britstein wrote:
>
> Add debug prints when creating and destroying rte flows.
>
> Signed-off-by: Eli Britstein
> Reviewed-by: Oz Shlomo
> ---
> lib/netdev-dpdk.c | 29 +++
> lib/netdev-offload-dpdk-flow.c| 168
>
On Wed, Nov 20, 2019 at 9:07 PM Eli Britstein wrote:
>
> Signed-off-by: Eli Britstein
> Reviewed-by: Oz Shlomo
> ---
> NEWS | 2 +
> lib/netdev-offload-dpdk-flow.c | 87
> --
> 2 files changed, 85 insertions(+), 4 deletions(-)
On Wed, Nov 20, 2019 at 9:06 PM Eli Britstein wrote:
>
> Currently, netdev datapath offload only accelerates the flow match
> sequence by associating a mark per flow. This series introduces the full
> offload of netdev datapath flows by having the HW also perform the flow
> actions.
>
> This
On Thu, Dec 5, 2019 at 2:38 PM Eli Britstein wrote:
>
>
> On 12/5/2019 9:55 AM, Sriharsha Basavapatna wrote:
> > Hi Eli,
> >
> > Any reason why tnl_pop is not included in this patch set ?
> >
> > Thanks,
> > -Harsha
>
> Currently offloading of vports does not work. Ilya works on that
> together
On Wed, Dec 4, 2019 at 8:55 PM Eli Britstein wrote:
>
>
> On 12/3/2019 5:19 PM, Sriharsha Basavapatna wrote:
> > On Wed, Nov 20, 2019 at 9:07 PM Eli Britstein wrote:
> >> Signed-off-by: Eli Britstein
> >> Reviewed-by: Oz Shlomo
> >> ---
> >> NEWS | 2 +
> >>
On Sun, Dec 8, 2019 at 3:08 PM Eli Britstein wrote:
>
>
> On 12/5/2019 6:46 PM, Sriharsha Basavapatna wrote:
> > On Wed, Dec 4, 2019 at 8:55 PM Eli Britstein wrote:
> >>
> >> On 12/3/2019 5:19 PM, Sriharsha Basavapatna wrote:
> >>> On Wed, Nov 20, 2019 at 9:07 PM Eli Britstein wrote:
>
On Tue, Feb 18, 2020 at 3:30 PM Eli Britstein wrote:
>
>
> On 2/10/2020 11:16 PM, Hemal Shah wrote:
> > Eli,
> >
> > There are some fundamental architecture issues (multi HW tables vs.
> > single HW table, use of rte_flow JUMP action for mapping ovs sw
> > tnl_pop action, etc.) that we need to
Hi Eli,
Thanks for sending this patchset. I have some questions about the
design, please see my comments below.
Thanks,
-Harsha
On Mon, Jan 20, 2020 at 8:39 PM Eli Britstein wrote:
>
> In the netdev datapath, packets arriving over a tunnel are processed by
> more than one flow. For example a
On Wed, Jan 8, 2020 at 11:34 PM Ilya Maximets wrote:
>
> On 08.01.2020 18:25, Eli Britstein wrote:
> >
> > On 1/8/2020 2:17 PM, Ilya Maximets wrote:
> >> On 19.12.2019 12:54, Eli Britstein wrote:
> >>> Flows that are offloaded via DPDK can be partially offloaded (matches
> >>> only) or fully
For flows that offload partial actions in egress direction,
provide the right netdev to fetch statistics.
Signed-off-by: Sriharsha Basavapatna
---
lib/dpif-netdev.c | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index
This patch refactors dp_netdev_flow_offload_put() to prepare for
more changes to support partial action offload, in subsequent patches.
- Add a wrapper function to allocate flow-mark.
- Move netdev_ports_get() to before flow-mark allocation.
Signed-off-by: Sriharsha Basavapatna
---
In this patch, we support offloading of VXLAN_ENCAP action for a vhost-user
port (aka "partial-action-offload"). At the time of offloading the flow, we
determine if the flow can be offloaded to an egress device, if the input
port is not offload capable such as a vhost-user port. We then offload
In this patch we check if action processing (apart from OUTPUT action),
should be skipped for a given dp_netdev_flow. Specifically, we check if
the action is TNL_PUSH and if it has been offloaded to HW, then we do not
push the tunnel header in SW. The datapath only executes the OUTPUT action.
The
Hi,
This RFC patchset extends the "Partial HW acceleration" mode to offload
part of the action processing to HW, instead of offloading just lookup
(MARK/RSS), for "vhost-user" ports. This is referred to as "Partial Action
Offload". This mode does not require SRIOV/switchdev configuration. In this
On Wed, May 6, 2020 at 5:56 PM Eli Britstein wrote:
>
>
> On 4/24/2020 11:23 AM, Sriharsha Basavapatna wrote:
> > In this patch we check if action processing (apart from OUTPUT action),
> > should be skipped for a given dp_netdev_flow. Specifically, we check if
> > the action is TNL_PUSH and if
On Wed, May 6, 2020 at 5:56 PM Eli Britstein wrote:
>
>
> On 4/24/2020 11:23 AM, Sriharsha Basavapatna wrote:
>
> In this patch, we support offloading of VXLAN_ENCAP action for a vhost-user
> port (aka "partial-action-offload"). At the time of offloading the flow, we
> determine if the flow can
In this patch we check if action processing (apart from OUTPUT action),
should be skipped for a given dp_netdev_flow. Specifically, we check if
the action is TNL_PUSH and if it has been offloaded to HW, then we do not
push the tunnel header in SW. The datapath only executes the OUTPUT action.
The
Hi,
This RFC patchset extends the "Partial HW acceleration" mode to offload
part of the action processing to HW, instead of offloading just lookup
(MARK/RSS), for "vhost-user" ports. This is referred to as "Partial Action
Offload". This mode does not require SRIOV/switchdev configuration. In this
This patch refactors dp_netdev_flow_offload_put() to prepare for
more changes to support partial action offload, in subsequent patches.
- Add a wrapper function to allocate flow-mark.
- Move netdev_ports_get() to before flow-mark allocation.
Signed-off-by: Sriharsha Basavapatna
---
In this patch, we support offloading of VXLAN_ENCAP action for a vhost-user
port (aka "partial-action-offload"). At the time of offloading the flow, we
determine if the flow can be offloaded to an egress device, if the input
port is not offload capable such as a vhost-user port. We then offload
This patch adds a function to determine if a given netdev belongs to the
dpdk-vhost class, using the netdev_class specific data.
Signed-off-by: Sriharsha Basavapatna
---
lib/netdev-dpdk.c | 5 +
lib/netdev-dpdk.h | 1 +
2 files changed, 6 insertions(+)
diff --git a/lib/netdev-dpdk.c
For flows that offload partial actions in egress direction,
provide the right netdev to fetch statistics.
Signed-off-by: Sriharsha Basavapatna
---
lib/dpif-netdev.c | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index
Hi Eli,
Thanks for your comments, please see my response inline, in the
respective patches.
Thanks,
-Harsha
On Wed, May 6, 2020 at 5:56 PM Eli Britstein wrote:
>
>
> On 4/24/2020 11:23 AM, Sriharsha Basavapatna wrote:
> > This patch refactors dp_netdev_flow_offload_put() to prepare for
> >
On Wed, May 6, 2020 at 5:56 PM Eli Britstein wrote:
>
>
> On 4/24/2020 11:23 AM, Sriharsha Basavapatna wrote:
> > For flows that offload partial actions in egress direction,
> > provide the right netdev to fetch statistics.
> >
> > Signed-off-by: Sriharsha Basavapatna
> > ---
> >
On Wed, May 20, 2020 at 3:38 PM Eli Britstein wrote:
>
>
> On 5/20/2020 12:12 PM, Sriharsha Basavapatna wrote:
> > On Mon, May 18, 2020 at 9:10 PM Eli Britstein wrote:
> >> For DPDK, there is the RAW_ENCAP attribute which gets raw buffer of the
> >> encapsulation header. For specific protocol,
On Mon, May 18, 2020 at 9:10 PM Eli Britstein wrote:
>
> The function of adding patterns by requested matches checks that it
> consumed all the required matches, and err if not. This nullify the
> purpose of the validation function. Future supported matches will only
> change the pattern parsing
On Mon, May 18, 2020 at 9:10 PM Eli Britstein wrote:
>
> Signed-off-by: Eli Britstein
> Reviewed-by: Roni Bar Yanai
> ---
> Documentation/howto/dpdk.rst | 1 +
> NEWS | 2 ++
> lib/netdev-offload-dpdk.c| 35 +++
> 3 files changed,
On Mon, May 18, 2020 at 9:10 PM Eli Britstein wrote:
>
> Add support for IPv6 pattern matching for offloading flows.
>
> Signed-off-by: Eli Britstein
> Reviewed-by: Roni Bar Yanai
> ---
> Documentation/howto/dpdk.rst | 2 +-
> NEWS | 1 +
> lib/netdev-offload-dpdk.c
Parse VLAN PUSH/POP OVS datapath actions and add respective RTE actions.
Signed-off-by: Sriharsha Basavapatna
---
lib/netdev-offload-dpdk.c | 34 ++
1 file changed, 34 insertions(+)
diff --git a/lib/netdev-offload-dpdk.c b/lib/netdev-offload-dpdk.c
index
On Sun, May 24, 2020 at 2:52 PM Tonghao Zhang wrote:
>
> On Sun, May 24, 2020 at 3:38 PM Sriharsha Basavapatna via dev
> wrote:
> >
> > Parse VLAN PUSH/POP OVS datapath actions and add respective RTE actions.
> One question, if pop vlan in the hw, then process it
On Sun, Sep 6, 2020 at 5:51 PM Eli Britstein wrote:
>
> ping
>
> On 7/30/2020 1:58 PM, Eli Britstein wrote:
> > From: Lei Wang
> >
> > Struct match has the tunnel values/masks in
> > match->flow.tunnel/match->wc.masks.tunnel.
> > Load actions such as load:0xa566c10->NXM_NX_TUN_IPV4_DST[],
> >
On Mon, Oct 12, 2020 at 4:48 PM Eli Britstein wrote:
>
>
> On 10/12/2020 10:20 AM, Sriharsha Basavapatna wrote:
> > On Sun, Sep 6, 2020 at 5:51 PM Eli Britstein wrote:
> >> ping
> >>
> >> On 7/30/2020 1:58 PM, Eli Britstein wrote:
> >>> From: Lei Wang
> >>>
> >>> Struct match has the tunnel
On Mon, Oct 12, 2020 at 4:43 PM Eli Britstein wrote:
>
>
> On 10/12/2020 11:45 AM, Sriharsha Basavapatna wrote:
> > On Sun, Sep 6, 2020 at 5:52 PM Eli Britstein wrote:
> >> ping
> >>
> >> On 7/30/2020 4:52 PM, Eli Britstein wrote:
> >>> In case of a flow modification, preserve the HW statistics
On Sun, Sep 6, 2020 at 5:52 PM Eli Britstein wrote:
>
> ping
>
> On 7/30/2020 4:52 PM, Eli Britstein wrote:
> > In case of a flow modification, preserve the HW statistics of the old HW
> > flow to the new one.
> >
> > Signed-off-by: Eli Britstein
> > Reviewed-by: Gaetan Rivet
Update fixes: tag
On Wed, Oct 14, 2020 at 8:29 PM Mark Gray wrote:
>
> This seems to break tests for me? Did you run a "make check"? I didn't
> apply the whole series because patch 2 doesn't apply so maybe that is
> needed for the tests to pass.
I rebased patch-2 and it now applies fine. But I'm seeing errors
On Mon, Oct 19, 2020 at 7:59 PM Ilya Maximets wrote:
>
> On 10/18/20 9:10 AM, Eli Britstein wrote:
> > Struct match has the tunnel values/masks in
> > match->flow.tunnel/match->wc.masks.tunnel.
> > Load actions such as load:0xa566c10->NXM_NX_TUN_IPV4_DST[],
> > load:0xbba->NXM_NX_TUN_ID[] are
On Fri, Jul 10, 2020 at 5:37 PM Sriharsha Basavapatna
wrote:
>
> Hi,
>
> This patchset fixes some issues found during netdev-offload-dpdk testing.
>
> Patch-1: Initialize rte 'transfer' attribute for mark/rss offload.
Patch-1 is not needed (discussed in an earlier thread).
> Patch-2: Pass L4
On Tue, Oct 20, 2020 at 2:40 PM Ilya Maximets wrote:
>
> On 10/20/20 10:51 AM, Sriharsha Basavapatna wrote:
> > On Mon, Oct 19, 2020 at 7:59 PM Ilya Maximets wrote:
> >>
> >> On 10/18/20 9:10 AM, Eli Britstein wrote:
> >>> Struct match has the tunnel values/masks in
> >>>
On Mon, Oct 12, 2020 at 7:57 PM Eli Britstein wrote:
>
> In case of a flow modification, preserve the HW statistics of the old HW
> flow to the new one.
>
> Fixes: 3c7330ebf036 ("netdev-offload-dpdk: Support offload of output action.")
> Signed-off-by: Eli Britstein
> Reviewed-by: Gaetan Rivet
On Mon, Oct 19, 2020 at 7:59 PM Ilya Maximets wrote:
>
> On 10/18/20 9:10 AM, Eli Britstein wrote:
> > Struct match has the tunnel values/masks in
> > match->flow.tunnel/match->wc.masks.tunnel.
> > Load actions such as load:0xa566c10->NXM_NX_TUN_IPV4_DST[],
> > load:0xbba->NXM_NX_TUN_ID[] are
1 - 100 of 239 matches
Mail list logo