On Fri, Dec 15, 2023 at 4:28 PM wrote:
>
> From: Kiran Kumar K
>
> While creating a RSS rule, if no RSS types are specified,
> use RSS types from dev config.
>
> Fixes: bc778a17fa46 ("net/cnxk: support flow RSS")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Kiran Kumar K
> Reviewed-by: Satheesh Paul
On Fri, 15 Dec 2023 17:26:24 +
Euan Bourke wrote:
> A recent thread on the mailing list[1] discussed corelist and coremask
> parsing and the idea of a new library dedicated to command line parsing
> was mentioned[2]. This patchset adds the library, along with the new
> APIs, and edits the exi
Update to the l3fwd-power example application to call the arg parser
library for its 'combined core string parser' instead of implementing its
own corelist parser. The default_type passed into the function call is
a corelist.
Signed-off-by: Euan Bourke
Acked-by: David Hunt
---
examples/l3fwd-po
Update to the eventdev_pipeline example application to call the arg parser
library for its 'combined core string parser' instead of implementing its
own coremask parser. The default_type passed into the function call is
a coremask.
Signed-off-by: Euan Bourke
---
examples/eventdev_pipeline/main.c
Two new functions, the first is a 'heuristic parser' which examines a
string describing a set of cores and determines based off heuristics
whether its a coremask or a corelist.
Second is a 'combined parser' which calls the first function and then
based off the returned value will call the relevant
Switched the dlb2 driver to call the new arg parsing library instead of
eal for coremask parsing, and updated the resource probe function to
support the changed formatting of the API.
Signed-off-by: Euan Bourke
---
drivers/event/dlb2/dlb2_priv.h | 4 +-
drivers/event/dlb2/pf/base/dl
Updates to the parse service cores functions in EAL to call the arg
parser API instead of implementing its own versions.
Signed-off-by: Euan Bourke
---
lib/eal/common/eal_common_options.c | 171 +++-
1 file changed, 41 insertions(+), 130 deletions(-)
diff --git a/lib/eal
Update to eal functions relating to corelist and coremask
parsing to support the new arg parsing library. Functions
now call the API instead of implementing their own version.
Signed-off-by: Euan Bourke
---
lib/eal/common/eal_common_options.c | 114
lib/eal/meson.bui
Add new coremask parsing API. This API behaves similarly
to the corelist parsing API, taking a coremask string, a cores
array and a cores_len int. Parsing the coremask string, filling
its values into the cores array up to cores_len.
The API also returns a 'count' which corresponds to the total num
Add a new library to make it easier for eal and other libraries to parse
command line arguments.
The first function in this library is one to parse a corelist string
into an array of individual core ids. The function will then return the
total number of cores described in the corelist.
Signed-off
A recent thread on the mailing list[1] discussed corelist and coremask
parsing and the idea of a new library dedicated to command line parsing
was mentioned[2]. This patchset adds the library, along with the new
APIs, and edits the existing EAL, DLB2 driver and some example
application functions to
On 12/15/2023 1:55 PM, Sivaramakrishnan, VenkatX wrote:
> Hi Ferruh,
>
>> -Original Message-
>> From: Ferruh Yigit
>> Sent: Friday, December 15, 2023 7:22 PM
>> To: Sivaramakrishnan, VenkatX ; Hemant
>> Agrawal ; Sachin Saxena
>> ; Zyta Szpak ; Liron Himi
>> ; Chaoyong He ; Gagandeep
>> S
The support of BPF in TAP device was problematic since the
API to BPF in kernel is not stable and keeps changing.
Since the RTE_FLOW_ACTION_TYPE_RSS is the only part using this
remove support for kernel side flow remapping.
Signed-off-by: Stephen Hemminger
---
doc/guides/nics/tap.rst
15/12/2023 14:44, Ferruh Yigit:
> On 12/14/2023 5:26 PM, Stephen Hemminger wrote:
> > On Thu, 14 Dec 2023 17:18:25 +
> > Ori Kam wrote:
> >
> Since encap groups number of different 5 tuples together, if HW doesn’t
> know
> how to RSS
> based on the inner application will
With the original 'use_cni' implementation, (using a
hardcoded socket rather than a configurable one),
if a DPDK pod is requesting multiple net devices
and these devices are from different pools, then
the container attempts to mount all the netdev UDSes
in the pod as /tmp/afxdp.sock. Which means th
On 15/12/2023 12:24, Ferruh Yigit wrote:
On 12/15/2023 11:15 AM, Maryam Tahhan wrote:
Add a new feature item that references the AF_XDP
DP integration.
Signed-off-by: Maryam Tahhan
Hi Maryam,
Better to have documentation (including release note update) and code in
same patch.
If there will
Thank you Maryam.
Patch looks fine we can go forward with merge.
One small comment on the commit message.
Regards,
Shibin
> -Original Message-
> From: Loftus, Ciara
> Sent: Friday, December 15, 2023 11:52 AM
> To: Tahhan, Maryam ; ferruh.yi...@amd.com;
> step...@networkplumber.org; lih
Hi Ferruh,
> -Original Message-
> From: Ferruh Yigit
> Sent: Friday, December 15, 2023 7:22 PM
> To: Sivaramakrishnan, VenkatX ; Hemant
> Agrawal ; Sachin Saxena
> ; Zyta Szpak ; Liron Himi
> ; Chaoyong He ; Gagandeep
> Singh ; Jerin Jacob ; Maciej Czekaj
>
> Cc: dev@dpdk.org; Power, Cia
On 12/15/2023 1:38 PM, Sivaramakrishnan Venkat wrote:
> Incorrect ptypes list causes buffer overflow for Address Sanitizer
> run. The last element in the ptypes lists to be "RTE_PTYPE_UNKNOWN"
> for rte_eth_dev_get_supported_ptypes().
> In rte_eth_dev_get_supported_ptypes(),the loop iterates until
On 12/14/2023 5:26 PM, Stephen Hemminger wrote:
> On Thu, 14 Dec 2023 17:18:25 +
> Ori Kam wrote:
>
Since encap groups number of different 5 tuples together, if HW doesn’t
know
how to RSS
based on the inner application will not be able to get any distribution of
>
Incorrect ptypes list causes buffer overflow for Address Sanitizer
run. The last element in the ptypes lists to be "RTE_PTYPE_UNKNOWN"
for rte_eth_dev_get_supported_ptypes().
In rte_eth_dev_get_supported_ptypes(),the loop iterates until it
finds "RTE_PTYPE_UNKNOWN" to detect last element of the pty
Sometimes it can be nice to have autotests which can take a parameter,
or can be tweaked in some ways, e.g. adjust the number of iterations, or
the burst size used in the test. Currently there is no way to do so -
all test parameters are hardcoded, which makes sense for a generic
regression test to
On 12/14/2023 10:24 AM, Chaoyong He wrote:
> From: Long Wu
>
> Set the representor array to NULL to avoid that close interface
> does not free some resource.
>
> Fixes: a135bc1644d6 ("net/nfp: fix resource leak for flower firmware")
> Cc: chaoyong...@corigine.com
> Cc: sta...@dpdk.org
>
> Signe
On 12/15/2023 11:15 AM, Maryam Tahhan wrote:
> Add a new feature item that references the AF_XDP
> DP integration.
>
> Signed-off-by: Maryam Tahhan
>
Hi Maryam,
Better to have documentation (including release note update) and code in
same patch.
If there will be a new version please combine bo
On 12/14/2023 10:58 AM, Michael Baum wrote:
> Add support for matching random value using new "rte_flow_item_random"
> structure.
> This random value is not based on the packet data/headers.
> Application shouldn't assume that this value is kept during the lifetime
> of the packet.
>
> v2:
> - Ra
>
> With the original 'use_cni' implementation, (using a
> hardcoded socket rather than a configurable one),
> if a DPDK pod is requesting multiple net devices
> and these devices are from different pools, then
> the container attempts to mount all the netdev UDSes
> in the pod as /tmp/afxdp.sock.
Hello,
I am solving the implementation of qemu-kvm with dpdk through vhost-user +
VPP. I have a problem with the fact that we have AMD epic servers with 4
numa nodes deployed in production. So 8 numa nodes for the entire server.
If we start virtualization and allow the guest to allocate a
Add a new feature item that references the AF_XDP
DP integration.
Signed-off-by: Maryam Tahhan
---
doc/guides/rel_notes/release_24_03.rst | 7 +++
1 file changed, 7 insertions(+)
diff --git a/doc/guides/rel_notes/release_24_03.rst
b/doc/guides/rel_notes/release_24_03.rst
index 6f8ad27808..
With the original 'use_cni' implementation, (using a
hardcoded socket rather than a configurable one),
if a DPDK pod is requesting multiple net devices
and these devices are from different pools, then
the container attempts to mount all the netdev UDSes
in the pod as /tmp/afxdp.sock. Which means th
Thanks Ciara, I missed the af_xdp.rst :(
Will update and push a v5 shortly
On 15/12/2023 09:55, Loftus, Ciara wrote:
Thanks for the latest patch Maryam. I have one minor suggestion inline.
Also, there are still some references to "use_cni" in af_xdp.rst which should
be removed/replaced with u
Add raw pattern support to VF, including FDIR/RSS flow rule.
This patch is based on DPDK v20.11.9
[ece54855816f1d03ef8ae08dedcb02318a97f3fb], for customer cherry-pick.
Signed-off-by: Mingjin Ye
---
drivers/common/iavf/virtchnl.h | 38 +-
drivers/net/iavf/iavf.h | 13 +-
d
Thanks for the latest patch Maryam. I have one minor suggestion inline.
Also, there are still some references to "use_cni" in af_xdp.rst which should
be removed/replaced with uds_path.
Once that's done I think the patch should be good to go. Perhaps also consider
adding a note to the release note
Hi Ferruh,
On 2023/12/15 17:24, Ferruh Yigit wrote:
> On 12/15/2023 12:43 AM, fengchengwen wrote:
>> Hi TB,
>>
>> On 2023/12/14 21:39, Mcnamara, John wrote:
>>> Release status meeting minutes 2023-12-14
>>>
>>> =
>>>
>>
>> ...
>>
>>>
>>> * main
>>>
>>> * V
On 12/15/2023 12:43 AM, fengchengwen wrote:
> Hi TB,
>
> On 2023/12/14 21:39, Mcnamara, John wrote:
>> Release status meeting minutes 2023-12-14
>>
>> =
>>
>
> ...
>
>>
>> * main
>>
>> * V1 date moved to December 29th
>>
>> * New branch mirroring to Gi
Added portforward usecase.In this usecase
packets received Rx port is forwarded to
respective Tx port.
Signed-off-by: Rakesh Kudurumalla
---
app/graph/ethdev.c | 13 ++
app/graph/ethdev.h | 1 +
app/graph/examples/l2fwd.cli |
Adds a txport to forward packet for every rxport
Mapping will be used to forward packets to txport
received on rxport
Following commands are exposed:
- ethdev forward "
Signed-off-by: Rakesh Kudurumalla
---
app/graph/cli.c | 1 +
app/graph/ethdev.c | 63 +
By default all packets received on ethdev_rx node
is forwarded to pkt_cls node.This patch provides
library support to add a new node as next node to
ethdev_rx node and forward packet to new node from
rx node.
Signed-off-by: Rakesh Kudurumalla
---
V5: Addressed comments
Handled negative test c
37 matches
Mail list logo
