Hi Guy
> -Original Message-
> From: g...@marvell.com
> Sent: Tuesday, October 20, 2020 11:16 AM
> dev@dpdk.org; jer...@marvell.com
> Subject: [PATCH v1] regexdev: add OOS capability definition
>
> From: Guy Kaneti
>
> Add out of order scan capability to check PMD
> support for OOS.
>
From: Guy Kaneti
configure qp with OOS according to device capabilities
returned from rte_regexdev_info_get.
Signed-off-by: Guy Kaneti
---
app/test-regex/main.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/app/test-regex/main.c b/app/test-regex/main.c
index e6080b44b.
Hi Thomas,
This issue has already been fixed, see:
http://mails.dpdk.org/archives/dev/2020-October/185921.html
It has been integrated, Long was able to test it and confirm it fixed
this issue.
On 20/10/20 00:36 +0200, Thomas Monjalon wrote:
> Fixing Gaetan's address
>
> 20/10/2020 00:33, Thomas
From: Ivan Malov
The patch adds APIs for client drivers to initialise / finalise
MAE-specific context in NIC control structure. The context
itself will be used by the follow-up patches to store
supportive data for library-internal consumers.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybch
From: Ivan Malov
These actions affect MAE supplementary resources which are libefx-internal.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/efsys.h | 2 +-
drivers/net/sfc/sfc_mae.c | 17 +
2 files changed,
Transfer rules for Alveo SN1000 SmartNICs support matching on various
inner and outer packet headers, traffic source items like PORT_ID,
PHY_PORT, PF and VF and actions to route traffic to destination
(PORT_ID, PHY_PORT, PF, VF or DROP), MARK, FLAG and apply VLAN
push/pop transformations.
Andrew
From: Ivan Malov
Add an API for client drivers to query the engine limits.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 10 +++
drivers/common/sfc_efx/base/efx_impl.h| 1 +
drivers/common/sfc_
From: Ivan Malov
Add a stub for MAE attach / detach path and introduce MAE-specific context.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/meson.build | 1 +
drivers/net/sfc/sfc.c | 8 ++
drivers/net/sfc/sfc.h | 2 +
From: Ivan Malov
Riverhead boards maintain support for MAE, a low-level Match-Action Engine.
The feature is documented in SF-122526-TC.
The new field will help client drivers to test
NIC support for MAE status.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreto
From: Ivan Malov
Define the corresponding specification structure and
make the code identify MAE rules by testing transfer
attribute presence. Also, add a priority level check.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst
Hi Guy,
> -Original Message-
> From: g...@marvell.com
> Sent: Tuesday, October 20, 2020 12:10 PM
> Subject: [PATCH v1] app/regex: configure qp according to device capabilities
>
> From: Guy Kaneti
>
> configure qp with OOS according to device capabilities
> returned from rte_regexdev
From: Ivan Malov
MAE has restrictions on what kind of mask a particular field can have in a
match specification. Add an API for client drivers to check specifications.
The patch defines a field description list, whilst the list itself is
left empty. This is to provide a general idea of how field
From: Ivan Malov
Add pattern parsing stub, define and implement flow cleanup method.
The latter is needed to free any dynamic structures allocated
during flow parsing.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/sfc_flow.c | 41
From: Ivan Malov
The new method is needed to make sure that a flow being
validated will have a chance to be accepted by the FW.
MAE-specific implementation of the method should
compare the class of a rule being validated with
the corresponding classes of active rules, and,
if no matches found, ma
From: Ivan Malov
An MAE rule is a function of match criteria and a priority. The said match
criteria have to be provided using "mask-value pairs" packing format which
on its own should not be exposed to client drivers. The latter have to use
a functional interface of sorts in order to generate a
From: Ivan Malov
Add MCDI-compatible enumeration for these fields and
provide necessary mappings for them to be inserted
directly into mask-value pairs buffer.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 3 ++
From: Ivan Malov
Add support for this flow item to MAE-specific RTE flow implementation.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 4 +++
drivers/net/sfc/sfc_mae.c | 69 +
drive
From: Ivan Malov
The action handler will use MAE action DELIVER with
MPORT of a given physical port.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 4
drivers/net/sfc/sfc_mae.c | 40 +++
From: Ivan Malov
Validate the match specification resulting from pattern
parsing within MAE backend in RTE flow implementation.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/sfc_mae.c | 8
1 file changed, 8 insertions(+)
dif
MAE stands for Match-Action-Engine and will be used to
support rte_flow API transfer rules.
Signed-off-by: Andrew Rybchenko
---
drivers/common/sfc_efx/base/efx_regs_mcdi.h | 2232 +++
1 file changed, 2232 insertions(+)
diff --git a/drivers/common/sfc_efx/base/efx_regs_mcdi.h
b/
From: Ivan Malov
>From MAE standpoint, a flow rule belongs to some class. Field capabilities
advertised by the FW provide a hint on whether changing a particular match
field value or its mask will affect the class of the rule. A client driver
can make use of the concept of a class by comparing a
From: Ivan Malov
The patch adds APIs for action set allocation / release.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 20 +++
drivers/common/sfc_efx/base/efx_mae.c | 128 ++
.
From: Ivan Malov
Exercise action set allocation / release and action rule
insertion / removal in order to let flow API callers
actually get created flows functioning.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/net/sfc/sfc_flow.c | 5 +-
From: Ivan Malov
Add support for this flow item to MAE-specific RTE flow implementation.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 2 +
drivers/net/sfc/sfc_mae.c | 106
2 file
From: Ivan Malov
MAE supports stripping two tags, so this action can
be requested once or twice.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 5 ++
drivers/common/sfc_efx/base/efx_impl.h| 7 +
From: Ivan Malov
MAE supports pushing two tags, so this action can
be requested once or twice.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 7 ++
drivers/common/sfc_efx/base/efx_impl.h| 12 +++
From: Ivan Malov
The engine is only able to carry out chosen actions on matching packets in
a strict order. No MCDI exists to identify supported actions and the order.
Still, the definition of the latter is available from the FW documentation.
The general idea is to define an action specificatio
From: Ivan Malov
If parsing a flow results in an action set specification
identical to an already existing one, duplication will
be avoided by reusing the list entry of the latter.
Using an attach helper and a reference counter
is meant to serve the said purpose.
Signed-off-by: Ivan Malov
Signe
From: Ivan Malov
Add an API for setting mask-value pairs in a match specification structure
and add support for MAE field INGRESS_PORT of type PPORT.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 48 ++
From: Ivan Malov
A group of actions (OF_PUSH_VLAN, OF_VLAN_SET_VID and
OF_VLAN_SET_PCP) maps to MAE action VLAN_PUSH.
This action group is supported only for rules which have transfer
attribute, and can be requested once or twice per a rule.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybc
From: Ivan Malov
Introduce a mechanism for adding actions to an action set and
add support for DELIVER action.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 11 +-
drivers/common/sfc_efx/base/efx_impl.
From: Ivan Malov
The action handler will use MAE action FLAG.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 2 ++
drivers/net/sfc/sfc_mae.c | 5 +
2 files changed, 7 insertions(+)
diff --git a/doc/guides/nics/sfc_
From: Ivan Malov
Add APIs for action rule insert / remove operations.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 29
drivers/common/sfc_efx/base/efx_mae.c | 150 ++
.../
From: Ivan Malov
There are MAE actions which do not have uniform counterparts
in terms of RTE flow. However, there are bundles of RTE flow
actions which can be considered as such counterparts.
Implement facilities to handle related RTE flow actions as
parts of a whole. These facilities will be u
From: Ivan Malov
This makes existing code clearer. Also, it will be used by a later patch.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 10 ++
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/
From: Ivan Malov
Effectively, the resulting action will be of type DELIVER, and
destination MPORT will be a properly constructed NULL value.
This will achieve the requested behaviour (no delivery).
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/gu
From: Ivan Malov
This action is supported only for rules which have transfer attribute,
and can be requested once or twice per a rule.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 2 ++
drivers/net/sfc/sfc_mae.c | 3 +
From: Ivan Malov
The action handler will use MAE action MARK.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 2 ++
drivers/net/sfc/sfc_mae.c | 12
2 files changed, 14 insertions(+)
diff --git a/doc/guides
From: Ivan Malov
This action can be added at any point before DELIVER.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 5 ++
drivers/common/sfc_efx/base/efx_impl.h| 8 +++
drivers/common/sfc_efx
From: Ivan Malov
The action handler will use MAE action DELIVER with MPORT of the PF/VF.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 4
drivers/net/sfc/sfc_mae.c | 34 ++
2 files
From: Ivan Malov
Add support for these flow items to MAE-specific RTE flow implementation.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 4 ++
drivers/net/sfc/sfc_mae.c | 126
dri
From: Ivan Malov
This action can be added at any point before DELIVER.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 6 ++
drivers/common/sfc_efx/base/efx_impl.h| 2 +
drivers/common/sfc_efx/b
From: Ivan Malov
The action handler will use MAE action DELIVER with MPORT
of the PCIe function associated with a given DPDK port ID.
The DPDK port must not relate to a different physical device.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guid
From: Ivan Malov
Add support for this flow item to MAE-specific RTE flow implementation.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 3 +
drivers/net/sfc/sfc_mae.c | 121
driver
From: Ivan Malov
The driver will need a means to figure out relationship between
RTE ethdev instances and underlying HW switch entities. For now,
use board serial number string as a unique HW switch identifier.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
From: Ivan Malov
Client drivers may need to request that matching traffic be dropped.
Add a dedicated API to support this. The API relies on action
DELIVER with properly constructed NULL MPORT argument.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
d
From: Ivan Malov
Add MCDI-compatible enumeration for these fields and
provide necessary mappings for them to be inserted
directly into mask-value pairs buffer.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 5 +
From: Ivan Malov
Riverhead boards can provide extended version information.
Implement facilities necessary to obtain it.
Add an API for querying board information.
A client driver may use this to discover which of its instances
relate to which physical boards, based on board serial number
persis
From: Ivan Malov
Add support for this flow item to MAE-specific RTE flow implementation.
The DPDK port must not relate to a different physical device.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 2 +
drivers/net/sfc/
From: Ivan Malov
PCIe functions have static MPORTs which can be utilised by
MAE rules as delivery destinations for matching traffic.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 14 +++
drivers/com
From: Ivan Malov
Add support for this flow item to MAE-specific RTE flow implementation.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 2 +
drivers/net/sfc/sfc_mae.c | 77 +
drivers
From: Ivan Malov
Refactor MCDI helper for version information and boot status
retrieval; it should comprise two dedicated helper functions.
A later patch will extend and reuse version retrieval helper.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/common/sfc_efx/base/
From: Ivan Malov
Match specification format and its size are not exposed to clients.
Provide an API to compare two match specifications.
A client would typically use this API to compare a match specification
of an outer rule being validated with match specifications of already
active outer rules
From: Ivan Malov
Let the client insert / remove outer rules.
Let the client refer to an inserted outer rule in a match
specification of type ACTION.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 21 ++
From: Ivan Malov
Let the client validate an outer match specification.
Let the client comprare classes of two outer match specifications.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx_impl.h | 3 +
drivers/common/sfc
From: Ivan Malov
Add MCDI-compatible enumeration for these fields and
provide necessary mappings for them to be inserted
directly into mask-value pairs buffer.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 4 ++
From: Ivan Malov
Add MCDI-compatible enumeration for this field and
provide necessary mappings for it to be inserted
directly into mask-value pairs buffer.
VNET_ID can be used to serve the following match fields:
rte_flow_item_vxlan.vni, rte_flow_item_geneve.vni,
rte_flow_item_nvgre.tni
Signed-
From: Ivan Malov
Add support for this flow item to MAE-specific RTE flow implementation.
In a pattern, a L2 item preceding an item VLAN must have
correct "type" ("inner_type") set depending on the total
number of VLAN tags (double-tagging is supported):
"pattern eth type is X / vlan / end",
X =
From: Ivan Malov
Add support for this flow item to MAE-specific RTE flow implementation.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 3 +
drivers/net/sfc/sfc_mae.c | 106
driver
From: Ivan Malov
Add support for this flow item to MAE-specific RTE flow implementation.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
doc/guides/nics/sfc_efx.rst | 2 ++
drivers/net/sfc/sfc_mae.c | 53 +
driver
Transfer rules support matching on various inner and outer packet
headers, traffic source items like PORT_ID, PHY_PORT, PF and VF and
actions to route traffic to destination (PORT_ID, PHY_PORT, PF, VF or
DROP), MARK, FLAG and apply VLAN push/pop transformations.
Signed-off-by: Andrew Rybchenko
--
From: Ivan Malov
Add support for flow items VXLAN, Geneve and NVGRE to
MAE-specific RTE flow implementation.
Having support for these items implies the ability to insert
so-called outer MAE rules and refer to them in MAE action rules.
The patch takes care of all necessary facilities to do that.
From: Ivan Malov
Add MCDI-compatible enumeration for these fields and
provide necessary mappings for them to be inserted
directly into mask-value pairs buffer.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 2 ++
From: Ivan Malov
A later patch will add support for RTE flow action PORT_ID
to MAE backend. The driver has to ensure that such actions
refer to RTE ethdev instances deployed on top of the same
physical device. Also, the driver needs a means to find
sibling RTE ethdev instances when parsing such a
From: Ivan Malov
Add MCDI-compatible enumeration for these fields and
provide necessary mappings for them to be inserted
directly into mask-value pairs buffer.
These fields are meant to comprise a so-called outer
match specification; provide necessary definitions.
Signed-off-by: Ivan Malov
Sig
From: Ivan Malov
Add MCDI-compatible enumeration for these fields and
provide necessary mappings for them to be inserted
directly into mask-value pairs buffer.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
Reviewed-by: Andy Moreton
---
drivers/common/sfc_efx/base/efx.h | 3 ++
From: Ivan Malov
MAE provides support for encapsulation. One needs to insert
a so-called outer rule, which can match outer packet fields,
to require that matching packets be parsed as tunnel frames
of a given type (VXLAN, Geneve, NVGRE). Then it is possible
to chain this rule with an action rule
When building kernel modules such as kni, the "config" directory is not
passed as a standard path in the EXTRA_CFLAGS value, meaning that the
rte_compatibility_defines.h is not found from rte_config.h. However, since
both headers are in the same directory, we can just use quotes rather than
angle-b
Hi Honnappa,
I will send out the next version of the patch soon.
Regards
Abhinandan
> -Original Message-
> From: Honnappa Nagarahalli
> Sent: Tuesday, October 20, 2020 3:34 AM
> To: Ananyev, Konstantin ; Gujjar, Abhinandan S
> ; dev@dpdk.org; Doherty, Declan
>
> Cc: jer...@marvell.com;
On Tue, 20 Oct 2020 11:48:15 +0300, Tal Shnaiderman wrote:
> Following the addition of the in_addr/in6_addr structs
> to in.h the header file must have stdint.h included
> for the definitions of the uint8_t/uint32_t types used
> within the new structs.
>
> Not having it could results in the follow
On 20/10/2020 10:11, Maxime Coquelin wrote:
>
>
> On 10/20/20 11:01 AM, Kevin Traynor wrote:
>> On 20/10/2020 08:16, Adrian Moreno wrote:
>>> If stat fails it means the backend must be vhost-user in server mode
>>>
>>> Bugzilla ID: 559
>>> Fixes: f908b22ea47a ("net/virtio: move backend type selec
https://bugs.dpdk.org/show_bug.cgi?id=559
xizhan4x (xix.zh...@intel.com) changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
On 10/20/20 11:38 AM, Kevin Traynor wrote:
> On 20/10/2020 10:11, Maxime Coquelin wrote:
>>
>>
>> On 10/20/20 11:01 AM, Kevin Traynor wrote:
>>> On 20/10/2020 08:16, Adrian Moreno wrote:
If stat fails it means the backend must be vhost-user in server mode
Bugzilla ID: 559
Fix
20/10/2020 10:37, Bruce Richardson:
> On Mon, Oct 19, 2020 at 06:58:32PM +0200, Thomas Monjalon wrote:
> > 15/10/2020 17:05, Bruce Richardson:
> > > Rather than specifying specific drivers in the driver directory to load,
> > > we
> > > can just pass in the whole driver directory to the "-d" EAL f
On 10/20/2020 10:00 AM, oulijun wrote:
在 2020/10/16 18:57, Ferruh Yigit 写道:
On 10/16/2020 11:04 AM, oulijun wrote:
在 2020/10/16 7:53, Ferruh Yigit 写道:
On 10/16/2020 12:21 AM, Ophir Munk wrote:
From: Ferruh Yigit
On 10/15/2020 1:41 PM, Lijun Ou wrote:
When start the testpmd, the pmd dri
This patchset resolves a coverity issue where the return value of
rte_eth_switch_domain_alloc() was not being checked and also removes
a use of rte_panic within ixgbe_pf_host_init().
---
v2: eth_dev->data->mac_addrs and eth_dev->data->hash_mac_addrs will now
be free'd if ixgbe_pf_host_init() fail
Remove the usage of rte_panic() within ixgbe_pf_host_init()
Signed-off-by: Conor Walsh
---
drivers/net/ixgbe/ixgbe_pf.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_pf.c b/drivers/net/ixgbe/ixgbe_pf.c
index 4e25b1e72e..833863af5a 100644
--- a
The return value of rte_eth_switch_domain_alloc() was not being checked
within ixgbe_pf_host_init() which caused a coverity issue. If the call
fails a warning is logged using PMD_INIT_LOG() and *vfinfo is free'd.
ixgbe_pf_host_init() now has a return value which is checked in
eth_ixgbe_dev_init()
On Tue, Oct 20, 2020 at 11:19 AM Bruce Richardson
wrote:
>
> When building kernel modules such as kni, the "config" directory is not
> passed as a standard path in the EXTRA_CFLAGS value, meaning that the
> rte_compatibility_defines.h is not found from rte_config.h. However, since
> both headers a
20/10/2020 10:34, Bruce Richardson:
> On Mon, Oct 19, 2020 at 11:04:54PM +0200, Thomas Monjalon wrote:
> > 19/10/2020 12:21, Bruce Richardson:
> > > On Fri, Oct 16, 2020 at 06:46:12PM +0200, David Marchand wrote:
> > > > On Fri, Oct 16, 2020 at 5:56 PM Bruce Richardson
> > > > wrote:
> > > > > > l
On Tue, Oct 20, 2020 at 12:04:56PM +0200, Thomas Monjalon wrote:
> 20/10/2020 10:34, Bruce Richardson:
> > On Mon, Oct 19, 2020 at 11:04:54PM +0200, Thomas Monjalon wrote:
> > > 19/10/2020 12:21, Bruce Richardson:
> > > > On Fri, Oct 16, 2020 at 06:46:12PM +0200, David Marchand wrote:
> > > > > On
> -Original Message-
> From: Su, Simei
> Sent: Tuesday, October 20, 2020 4:47 PM
> To: Zhang, Qi Z ; Yang, Qiming
> Cc: dev@dpdk.org; Allan, Bruce W ; Su, Simei
>
> Subject: [PATCH] net/ice/base: fix a refactor error
>
> This patch corrects an upper limit value in for loop.
>
> Fixe
16/10/2020 00:57, Stephen Hemminger:
> --- a/lib/librte_eal/rte_eal_version.map
> +++ b/lib/librte_eal/rte_eal_version.map
> - rte_get_master_lcore;
> + rte_get_main_lcore;
The .def must be updated as well.
I will fix while merging.
16/10/2020 00:57, Stephen Hemminger:
> As announced in 20.08 release notes, the terms for lcore role
> will change in 20.11 release.
>
> This is a project wide patchset to terminolgy used for DPDK lcore.
> The term "master" referring to lcore is replaced with "main" and
> "slave" lcore is replaced
> -Original Message-
> From: dev On Behalf Of Junyu Jiang
> Sent: Tuesday, October 20, 2020 4:51 PM
> To: dev@dpdk.org
> Cc: Yang, Qiming ; Su, Simei ;
> Jiang, JunyuX ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH] net/ice: fix SCTP RSS configuration
>
> This patch configured RSS for
> -Original Message-
> From: Rong, Leyi
> Sent: Monday, October 19, 2020 1:43 PM
> To: Zhang, Qi Z ; Yigit, Ferruh
> Cc: dev@dpdk.org; Rong, Leyi ; sta...@dpdk.org
> Subject: [PATCH v2] net/iavf: fix unchecked Tx cleanup error
>
> Coverity complains of unchecked return value warning o
On Tue, Oct 20, 2020 at 12:03 PM David Marchand
wrote:
>
> On Tue, Oct 20, 2020 at 11:19 AM Bruce Richardson
> wrote:
> >
> > When building kernel modules such as kni, the "config" directory is not
> > passed as a standard path in the EXTRA_CFLAGS value, meaning that the
> > rte_compatibility_def
On 10/20/20 11:55 AM, Maxime Coquelin wrote:
>
>
> On 10/20/20 11:38 AM, Kevin Traynor wrote:
>> On 20/10/2020 10:11, Maxime Coquelin wrote:
>>>
>>>
>>> On 10/20/20 11:01 AM, Kevin Traynor wrote:
On 20/10/2020 08:16, Adrian Moreno wrote:
> If stat fails it means the backend must be vh
On Sun, Oct 18, 2020 at 4:14 PM Andrew Rybchenko
wrote:
> diff --git a/doc/guides/rel_notes/deprecation.rst
> b/doc/guides/rel_notes/deprecation.rst
> index 18a748bc88..42eafb97af 100644
> --- a/doc/guides/rel_notes/deprecation.rst
> +++ b/doc/guides/rel_notes/deprecation.rst
> @@ -106,12 +106,6
On Sun, Oct 18, 2020 at 4:14 PM Andrew Rybchenko
wrote:
>
> The command uses FDIR filter information get API which
> is not supported any more.
>
> Signed-off-by: Andrew Rybchenko
Build error on parse_flexbytes():
FAILED: app/a172ced@@dpdk-testpmd@exe/test-pmd_cmdline.c.o
ccache clang -Iapp/a17
On Sun, Oct 18, 2020 at 4:10 PM Andrew Rybchenko
wrote:
>
> RTE flow API should be used for filtering.
- We still have some trace in testpmd documentation.
$ git grep set.port.*vf.*mac
doc/guides/testpmd_app_ug/testpmd_funcs.rst: testpmd> set port
(port_id) vf (vf_id) (mac_addr) \
- Do we hav
Add AVX512 support for ice PMD. This patch adds ice_rxtx_vec_avx512.c
to support ice AVX512 vPMD.
This patch aims to enable AVX512 on ice vPMD. Main changes are focus
on Rx path compared with AVX2 vPMD.
Signed-off-by: Leyi Rong
Signed-off-by: Bruce Richardson
---
drivers/net/ice/ice_rxtx.c
This patchset aims to support AVX512 vPMD on ice.
---
v3:
- Code rebased.
v2:
- No internal judgement when RTE_MACHINE_CPUFLAG_AVX512F is set in
meson.build.
- Add RSS hash parsing as default RXDID is set to #22.
- Use buf_iova instead of buf_physaddr, as buf_physaddr will be removed.
Leyi Ron
Support RSS hash parsing in AVX512 data path as the default
RXDID is set to #22, that means the RSS hash field locates
in the 2nd 16B of each Flex Rx descriptor.
Signed-off-by: Leyi Rong
---
drivers/net/ice/ice_rxtx_vec_avx512.c | 105 --
1 file changed, 98 insertions(+),
Optimize Tx path by using AVX512 instructions and vectorize the
tx free bufs process.
Signed-off-by: Leyi Rong
Signed-off-by: Bruce Richardson
---
drivers/net/ice/ice_rxtx.h| 4 +
drivers/net/ice/ice_rxtx_vec_avx512.c | 155 +-
drivers/net/ice/ice_rxtx_vec_
On 10/20/2020 7:11 AM, Xiaoyun Li wrote:
This example sets up a scenario that VXLAN packets can be received
by different PF queues based on VNID and each queue is bound to a VM
with a VNID so that the VM can receive its inner packets.
Usually, OVS is used to do the software encap/decap for VXLAN
This patch set makes vhost-vswitch be able to use vhost async APIs
for enqueue operations. Demonstrated how the application
leverage IOAT DMA channel with vhost async APIs.
We introduce two parameters to enable DMA acceleration for Tx
operations of queues:
-dma_type This parameter is used to speci
This patch is to add async vhost driver arguments parsing function
for CBDMA channel, DMA initiation function and args description.
The meson build file is changed to fix dependency problem. With
these arguments vhost device can be set to use CBDMA or CPU for
enqueue operation and bind vhost device
This patch is to implement vhost DMA operation callbacks for CBDMA
PMD and add vhost async data-path in vhost sample. With providing
callback implementation for CBDMA, vswitch can leverage IOAT to
accelerate vhost async data-path.
Signed-off-by: Cheng Jiang
---
examples/vhost/ioat.c | 100 ++
Add release notes for vhost async data path support in vhost sample.
Signed-off-by: Cheng Jiang
---
doc/guides/rel_notes/release_20_11.rst | 6 ++
1 file changed, 6 insertions(+)
diff --git a/doc/guides/rel_notes/release_20_11.rst
b/doc/guides/rel_notes/release_20_11.rst
index f8686a50d..8
101 - 200 of 395 matches
Mail list logo
