The flow counter support introduced by
commit 9a761de8ea14 ("net/mlx5: flow counter support") was intend to
work only with MLNX_OFED_4.3 as the upstream rdma-core
libraries were lack such support.
On rdma-core v19 the support for the flow counters was added but with
different user APIs, hence caus
On Thursday 12 July 2018 03:14 AM, Gaetan Rivet wrote:
This abstraction exists since the infancy of DPDK.
It needs to be fleshed out however, to allow a generic
description of devices properties and capabilities.
A device class is the northbound interface of the device, intended
for applications
Hi Ori,
Pls see my reply.
Hi Walter and Ferruh,
I need your voice :)
> -Original Message-
> From: Ori Kam [mailto:or...@mellanox.com]
> Sent: Thursday, July 12, 2018 13:58
> To: Xu, Rosen ; dev@dpdk.org
> Cc: Yigit, Ferruh ; sta...@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH] examples/flo
Hi,
PSB
> -Original Message-
> From: Xu, Rosen [mailto:rosen...@intel.com]
> Sent: Thursday, July 12, 2018 8:27 AM
> To: Ori Kam ; dev@dpdk.org
> Cc: Yigit, Ferruh ; sta...@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH] examples/flow_filtering: add rte_fdir_conf
> initialization
>
> Hi Ori,
Hi Xiaoyun,
> -Original Message-
> From: Li, Xiaoyun
> Sent: Wednesday, July 11, 2018 10:16 AM
> To: Zhang, Qi Z ; Lu, Wenzhuo
>
> Cc: dev@dpdk.org; Li, Xiaoyun ; sta...@dpdk.org
> Subject: [PATCH v2] app/testpmd: fix little perf drop with XL710
>
> There is about 1.8M perf drop with XL
Hi Stephen,
Thursday, July 12, 2018 12:08 AM, Stephen Hemminger:
> Subject: [BUG] mlx5 build failure on Debian testing
>
> I am seeing a build failure of MLX5 driver on Debian testing (buster) with
> dpdk.org master branch.
>
> This is using libibverbs-dev version 19.0-1
>
> The failure is:
> m
>-Original Message-
>From: De Lara Guarch, Pablo [mailto:pablo.de.lara.gua...@intel.com]
>Sent: 11 July 2018 18:56
>To: Verma, Shally
>Cc: dev@dpdk.org; Athreya, Narayana Prasad
>; Challa, Mahipal
>; Sahu, Sunila ; Sahu,
>Sunila ; Gupta, Ashish
>
>Subject: RE: [PATCH v2 4/5] compress/
Hi Ori,
examples/flow_filtering sample app fails on i40e [1] because i40e requires
explicit FDIR configuration.
But rte_flow in and hardware independent ways of describing flow-action, it
shouldn't require specific config options for specific hardware.
Is there any chance driver select the FDI
The device information cannot be gotten correctly before
the configuration is set. Because on some NICs the
information has dependence on the configuration.
Fixes: 3be82f5cc5e3 ("ethdev: support PMD-tuned Tx/Rx parameters")
Signed-off-by: Wenzhuo Lu
---
lib/librte_ethdev/rte_ethdev.c | 47 ++
Hi Rosen,
Why do the fdir_conf must be initialized?
What is the issue you are seeing?
Best,
Ori
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Rosen Xu
> Sent: Thursday, July 12, 2018 5:10 AM
> To: dev@dpdk.org
> Cc: rosen...@intel.com; ferruh.yi...@intel.co
> -Original Message-
> From: Li, Xiaoyun
> Sent: Wednesday, July 11, 2018 10:16 AM
> To: Zhang, Qi Z ; Lu, Wenzhuo
> Cc: dev@dpdk.org; Li, Xiaoyun ; sta...@dpdk.org
> Subject: [PATCH v2] app/testpmd: fix little perf drop with XL710
>
> There is about 1.8M perf drop with XL710. And it i
On 7/11/2018 11:46 PM, Stephen Hemminger wrote:
On Wed, 11 Jul 2018 18:41:50 +0800
Jeff Guo wrote:
As we know, hot plug is an importance feature, either use for the datacenter
device’s fail-safe, or use for SRIOV Live Migration in SDN/NFV. It could bring
the higher flexibility and continual
The workaround of BAR0 mapping does not work if BAR0 area is smaller
than page size (64KB in ppc). In addition, we no longer need the
workaround in recent Linux because VFIO allows MSIX mapping (*).
This fix is just to skip the workaround if BAR0 is smarller than a page.
(*): "vfio-pci: Allow mapp
The workaround of BAR0 mapping does not work if BAR0 area is smaller
than page size (64KB in ppc). In addition, we no longer need the
workaround in recent Linux because VFIO allows MSIX mapping (*).
This fix is just to skip the workaround if BAR0 is smarller than a page.
(*): "vfio-pci: Allow mapp
SPDK blobfs encountered a crash around rte_ring dequeues in ppc64.
It uses a single consumer and multiple producers for a rte_ring.
The problem was a load-load reorder in rte_ring_sc_dequeue_bulk().
The reordered loads happened on r->prod.tail in
__rte_ring_move_cons_head() (rte_ring_generic.h) an
Hi Yipeng,
I agree with you on RCU. It solves only part of the problem and
requires application involvement. Use of atomics is required to solve few more
problems.
Not solving the preemptible writer issue will change the behavior for existing
applications, is that ok?
I will submit a R
Rte_fdir_conf of rte_eth_conf should be initialized before
port initialization.
Fixes: 4a3ef59a10c8 ("examples/flow_filtering: add simple demo of flow API")
Cc: sta...@dpdk.org
Signed-off-by: Rosen Xu
---
examples/flow_filtering/main.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/e
> -Original Message-
> From: Guo, Jia
> Sent: Wednesday, July 11, 2018 6:42 PM
>
> When device be hotplug out, the pci resource will be released in kernel,
> the uio fd will disappear, and the irq will be released. At this time,
> if igb uio driver still try to access or release these res
Hi, Honnappa,
Thanks for the comment.
RCU can handle one of the currency issue (key deletion while lookup) that has
been discussed before, but we think it is not easy for RCU-alone to
protect reader from cuckoo path displacement. Also, RCU solution requires user
interaction.
We agree that the
Hi, Stephen,
You are correct and we understand that spinlock might be slightly faster than
counter based rwlock in this case. However, the counter based rwlock is the
exception path when TSX fails.
If performance of this exception path is a big concern, a more optimal
read-write lock scheme (e
v13:
- Since rte_eth_dev_attach/rte_eth_dev_detach will be deprecated,
so, modify the sample code to use rte_eal_hotplug_add and
rte_eal_hotplug_remove to attach/detach device.
v12:
- fix return value in eal_dev_hotplug_request_to_primary.
- add more error log in rte_eal_hotplug_add.
- fix ret
Update release notes for the new multi-process hotplug feature.
Signed-off-by: Qi Zhang
---
doc/guides/rel_notes/release_18_08.rst | 11 +++
1 file changed, 11 insertions(+)
diff --git a/doc/guides/rel_notes/release_18_08.rst
b/doc/guides/rel_notes/release_18_08.rst
index bc0124295..12
v13:
- Since rte_eth_dev_attach/rte_eth_dev_detach will be deprecated,
so, modify the sample code to use rte_eal_hotplug_add and
rte_eal_hotplug_remove to attach/detach device.
v12:
- fix return value in eal_dev_hotplug_request_to_primary.
- add more error log in rte_eal_hotplug_add.
- fix ret
The sample code demonstrates device (ethdev only) management
at a multi-process environment. The user can attach/detach a
device on primary process and see it is synced on secondary
process automatically.
How to start?
./hotplug_mp --proc-type=auto
Command Line Example:
>help
>list
/* attach a
Attach port from secondary should ignore devargs since the private
device is not necessary to support. Also previously, detach port on
a secondary process will mess primary process and cause the same
device can't be attached back again. A secondary process should use
rte_eth_release_port_private to
Attach port from secondary should ignore devargs since the private
device is not necessary to support. Also previously, detach port on
a secondary process will mess primary process and cause the same
device can't be attached back again. A secondary process should use
rte_eth_release_port_private to
Attach port from secondary should ignore devargs since the private
device is not necessary to support. Also previously, detach port on
a secondary process will mess primary process and cause the same
device can't be attached back again. A secondary process should use
rte_eth_release_port_private to
Attach port from secondary should ignore devargs since the private
device is not necessary to support. Also previously, detach port on
a secondary process will mess primary process and cause the same
device can't be attached back again. A secondary process should use
rte_eth_release_port_private to
Attach port from secondary should ignore devargs since the private
device is not necessary to support. Also previously, detach port on
a secondary process will mess primary process and cause the same
device can't be attached back again. A secondary process should use
rte_eth_release_port_private to
Attach port from secondary should ignore devargs since the private
device is not necessary to support. Also previously, detach port on
a secondary process will mess primary process and cause the same
device can't be attached back again. A secondary process should use
rte_eth_release_port_private to
Attach port from secondary should ignore devargs since the private
device is not necessary to support. Also previously, detach port on
a secondary process will mess primary process and cause the same
device can't be attached back again. A secondary process should use
rte_eth_release_port_private to
Attach port from secondary should ignore devargs since the private
device is not necessary to support. Also previously, detach port on
a secondary process will mess primary process and cause the same
device can't be attached back again. A secondary process should use
rte_eth_release_port_private to
Subroutine to unmap VFIO resource is shared by secondary and
primary, and it does not work on the secondary process. Since
for secondary process, it is not necessary to close interrupt
handler, set pci bus mastering and remove vfio_res from
vfio_res_list. So, the patch adds a dedicate function to h
This patch cover the multi-process hotplug case when a device
attach/detach request be issued from a secondary process
device attach on secondary:
a) secondary send sync request to the primary.
b) primary receive the request and attach the new device if
failed goto i).
c) primary forward attach
Attach port from secondary should ignore devargs since the private
device is not necessary to support. Also previously, detach port on
a secondary process will mess primary process and cause the same
device can't be attached back again. A secondary process should use
rte_eth_release_port_private to
Previously, detach port on a secondary process will mess primary
process and cause the same device can't be attached back again.
A secondary process should use rte_eth_release_port_private to
release a port.
Signed-off-by: Qi Zhang
---
drivers/net/i40e/i40e_ethdev.c | 2 ++
1 file changed, 2 ins
Previously, detach port on a secondary process will mess primary
process and cause the same device can't be attached back again.
A secondary process should use rte_eth_release_port_private to
release a port.
Signed-off-by: Qi Zhang
---
drivers/net/ixgbe/ixgbe_ethdev.c | 3 +++
1 file changed, 3
Add driver API rte_eth_release_port_private to support the
case when an ethdev need to be detached on a secondary process.
Local state is set to unused and shared data will not be reset
so the primary process can still use it.
Signed-off-by: Qi Zhang
Reviewed-by: Andrew Rybchenko
Acked-by: Remy
v13:
- Since rte_eth_dev_attach/rte_eth_dev_detach will be deprecated,
so, modify the sample code to use rte_eal_hotplug_add and
rte_eal_hotplug_remove to attach/detach device.
v12:
- fix return value in eal_dev_hotplug_request_to_primary.
- add more error log in rte_eal_hotplug_add.
- fix ret
We are going to introduce the solution to handle hotplug in
multi-process, it includes the below scenario:
1. Attach a device from the primary
2. Detach a device from the primary
3. Attach a device from a secondary
4. Detach a device from a secondary
In the primary-secondary process model, we ass
When use memcmp to compare two PCI address, sizeof(struct rte_pci_addr)
is 4 bytes aligned, and it is 8. While only 7 byte of struct rte_pci_addr
is valid. So compare the 8th byte will cause the unexpected result, which
happens when repeatedly attach/detach a device.
Fixes: c752998b5e2e ("pci: int
Clear vfio_group_fd is not necessary to involve any IPC.
Also, current IPC implementation for SOCKET_CLR_GROUP is not
correct. rte_vfio_clear_group on secondary will always fail,
that prevent device be detached correctly on a secondary process.
The patch simply removes all IPC related stuff in
rte_
On Wed, Jun 27, 2018 at 08:08:20PM +0200, Adrien Mazarguil wrote:
> This enables flow rules to match traffic coming from a different DPDK port
> ID associated with the device (PORT_ID pattern item), mainly for the
> convenience of applications that want to deal with a single port ID for all
> flow
On Wed, Jun 27, 2018 at 08:08:18PM +0200, Adrien Mazarguil wrote:
> This enables flow rules to explicitly match VLAN traffic (VLAN pattern
> item) and perform various operations on VLAN headers at the switch level
> (OF_POP_VLAN, OF_PUSH_VLAN, OF_SET_VLAN_VID and OF_SET_VLAN_PCP actions).
>
> Test
On Wed, Jun 27, 2018 at 08:08:16PM +0200, Adrien Mazarguil wrote:
> This enables flow rules to explicitly match supported combinations of
> Ethernet, IPv4, IPv6, TCP and UDP headers at the switch level.
>
> Testpmd example:
>
> - Dropping TCPv4 traffic with a specific destination on port ID 2:
>
On Wed, Jun 27, 2018 at 08:08:14PM +0200, Adrien Mazarguil wrote:
> This patch enables creation of rte_flow rules that direct matching traffic
> to a different port (e.g. another VF representor) or drop it directly at
> the switch level (PORT_ID and DROP actions).
>
> Testpmd examples:
>
> - Dire
On Wed, Jun 27, 2018 at 08:08:12PM +0200, Adrien Mazarguil wrote:
> Because mlx5 switch flow rules are configured through Netlink (TC
> interface) and have little in common with Verbs, this patch adds a separate
> parser function to handle them.
>
> - mlx5_nl_flow_transpose() converts a rte_flow r
> -Original Message-
> From: Andrew Rybchenko [mailto:arybche...@solarflare.com]
> Sent: Thursday, July 12, 2018 12:05 AM
> To: Zhang, Qi Z ; tho...@monjalon.net; Burakov,
> Anatoly
> Cc: Ananyev, Konstantin ; dev@dpdk.org;
> Richardson, Bruce ; Yigit, Ferruh
> ; Shelton, Benjamin H
> ;
On Wed, Jun 27, 2018 at 08:08:10PM +0200, Adrien Mazarguil wrote:
> With mlx5, unlike normal flow rules implemented through Verbs for traffic
> emitted and received by the application, those targeting different logical
> ports of the device (VF representors for instance) are offloaded at the
> swit
18/06/2018 14:32, Thomas Monjalon:
> A constructor is usually declared with RTE_INIT* macros.
> As it is a static function, no need to declare before its definition.
> The macro is used directly in the function definition.
>
> Signed-off-by: Thomas Monjalon
Rebased after crypto pull and applied
> Jiayu Hu (3):
> gso: support UDP/IPv4 fragmentation
> app/testpmd: enable UDP GSO in csum engine
> gso: update documents for UDP/IPv4 GSO
Applied, thanks
Add the --dev parameter to the EAL.
This new parameter takes a generic device declaration as argument.
It uses the new devargs parsing API.
Signed-off-by: Gaetan Rivet
---
lib/librte_eal/common/eal_common_devargs.c | 4 +++
lib/librte_eal/common/eal_common_options.c | 36 +++---
Process the class-specific arguments in a devargs.
This processing takes the form of setting the proper eth_dev fields when
relevant.
Signed-off-by: Gaetan Rivet
---
lib/librte_ethdev/eth_private.h | 5 +++
lib/librte_ethdev/rte_class_eth.c | 62 +++
lib/librte_eth
The eth device class can now parse a field name,
matching the eth_dev name with one passed as
"class=eth,name=xx"
Signed-off-by: Gaetan Rivet
---
lib/librte_ethdev/rte_class_eth.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/lib/librte_ethdev/rte_class_eth.c
b/lib/librte_e
A new interactive command is offered:
show device
This commands lists all rte_device element matching the device
description. e.g.:
show device bus=pci
show device bus=vdev
show device bus=vdev/class=eth
show device bus=vdev,driver=net_ring/class=eth
show device bus=vdev/class
Introduce the facility to process future PCI parameters.
Once the matching between PCI devices and devargs has been done, it is
possible to process each devargs. New parameters would have the PCI
device handle to work with when parsing the device (bus specific)
parameters.
Signed-off-by: Gaetan R
The new devargs format does not recognize a particular device name.
Each bus uses its specific format.
Process each devargs privately prior to attempting a bus scan.
Prepare them if they are using the new devargs format.
Signed-off-by: Gaetan Rivet
---
drivers/bus/vdev/vdev.c | 10 -
The new devargs format does not recognize a particular device name.
Each bus uses its specific format.
Instead of introducing a new bus API, process those devargs privately
for the moment. Prepare them for matching during scan against the
bus devices.
Signed-off-by: Gaetan Rivet
---
drivers/bus
Signed-off-by: Gaetan Rivet
---
lib/librte_ethdev/Makefile| 3 +-
lib/librte_ethdev/meson.build | 1 +
lib/librte_ethdev/rte_class_eth.c | 79 +++
3 files changed, 82 insertions(+), 1 deletion(-)
create mode 100644 lib/librte_ethdev/rte_class_eth.c
diff
Signed-off-by: Gaetan Rivet
---
drivers/bus/vdev/Makefile | 3 +-
drivers/bus/vdev/meson.build| 5 +++-
drivers/bus/vdev/vdev.c | 10 ---
drivers/bus/vdev/vdev_params.c | 51 +
drivers/bus/vdev/vdev_private.h | 26 +
5 files
The PCI bus can now parse a matching field "id" as follows:
"bus=pci,id=:00:00.0"
or
"bus=pci,id=00:00.0"
Signed-off-by: Gaetan Rivet
---
drivers/bus/pci/pci_params.c | 29 +++--
1 file changed, 27 insertions(+), 2 deletions(-)
diff --git a/driver
This iterator can be customized with a comparison function that will
trigger a stopping condition.
It can be leveraged to write several different iterators that have
similar but non-identical purposes.
It is private to librte_ethdev.
Signed-off-by: Gaetan Rivet
---
lib/librte_ethdev/Makefile
The vdev bus parses a field "driver", matching
a vdev driver name with one passed as follows:
"bus=vdev,driver="
Signed-off-by: Gaetan Rivet
---
drivers/bus/vdev/vdev_params.c | 15 +--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/drivers/bus/vdev/vdev_param
Signed-off-by: Gaetan Rivet
---
drivers/bus/pci/Makefile | 3 +-
drivers/bus/pci/meson.build | 6 +++-
drivers/bus/pci/pci_common.c | 3 +-
drivers/bus/pci/pci_params.c | 53
drivers/bus/pci/private.h| 25 +
5 files changed, 86 inse
This function can be used as a callback to
rte_kvargs_process.
This should reduce code duplication.
Signed-off-by: Gaetan Rivet
---
lib/librte_kvargs/rte_kvargs.c | 10 +
lib/librte_kvargs/rte_kvargs.h | 26
lib/librte_kvargs/rte_kvargs_versi
A device iterator allows iterating over a set of devices.
This set is defined by the two descriptions offered,
* rte_bus
* rte_class
Only one description can be provided, or both. It is not allowed to
provide no description at all.
Each layer of abstraction then performs a filter based on th
Use the iteration hooks in the abstraction layers to perform the
requested filtering on the internal device lists.
Signed-off-by: Gaetan Rivet
---
lib/librte_eal/common/eal_common_dev.c | 168
lib/librte_eal/common/include/rte_dev.h | 26
lib/librte_eal/rte_eal_ve
This abstraction exists since the infancy of DPDK.
It needs to be fleshed out however, to allow a generic
description of devices properties and capabilities.
A device class is the northbound interface of the device, intended
for applications to know what it can be used for.
It is conceptually jus
Parse a device description.
Split this description in their relevant part for each layers.
No dynamic allocation is performed.
Signed-off-by: Gaetan Rivet
---
lib/Makefile| 1 +
lib/librte_eal/bsdapp/eal/Makefile | 1 +
lib/librte_eal/common/eal_common_dev.c |
This function is private to the EAL.
It is used to parse each layers in a device description string,
and store the result in an rte_devargs structure.
Signed-off-by: Gaetan Rivet
---
lib/librte_eal/common/eal_common_devargs.c | 144
lib/librte_eal/common/eal_private.h
This macro adds symbols to the .fini section using the global
RTE priorities, to ensure consistency.
Signed-off-by: Gaetan Rivet
Acked-by: Shreyansh Jain
---
lib/librte_eal/common/include/rte_common.h | 23 ++
1 file changed, 23 insertions(+)
diff --git a/lib/librte_eal/com
Error logs in kvargs parsing should be better handled in components
calling the library.
This library must be as lean as possible.
Signed-off-by: Gaetan Rivet
Acked-by: Thomas Monjalon
---
lib/librte_kvargs/rte_kvargs.c | 22 +-
1 file changed, 5 insertions(+), 17 deletions
rte_devargs_parse becomes non-variadic,
rte_devargs_parsef becomes the variadic version, to be used to compose
device strings.
Signed-off-by: Gaetan Rivet
Acked-by: Shreyansh Jain
Acked-by: Thomas Monjalon
---
drivers/net/failsafe/failsafe_args.c| 2 +-
drivers/net/failsafe/failsafe_e
This library will be used by the EAL to parse parameters.
Signed-off-by: Gaetan Rivet
---
lib/Makefile | 3 +--
lib/librte_eal/meson.build| 1 +
lib/librte_kvargs/Makefile| 2 +-
lib/librte_kvargs/meson.build | 3 +++
lib/meson.build | 8 ++--
5 files c
This function permits defining additional terminating characters,
ending the parsing to arbitrary delimiters.
Signed-off-by: Gaetan Rivet
Acked-by: Thomas Monjalon
Acked-by: Shreyansh Jain
---
lib/Makefile | 1 +
lib/librte_kvargs/meson.build| 2 ++
li
This patchset introduces a new EAL API for querying devices,
filtered by arbitrary properties.
The following elements are introduced to this end:
* A new object, "rte_class", is used to describe
the device class abstraction layer (eth, crypto, ...).
* Both rte_bus and rte_class now offer a
Use the standard EAL logging functions in rte_devargs.
Signed-off-by: Gaetan Rivet
---
lib/librte_eal/common/eal_common_devargs.c | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/lib/librte_eal/common/eal_common_devargs.c
b/lib/librte_eal/common/eal_common_devargs.c
in
09/05/2018 18:22, Erik Gabriel Carrillo:
> Add APIs that allow an application to query and reset the attributes of
> a service lcore. Add one such new attribute, "loops", which is a
> counter that tracks the number of times the service core has looped in
> the service runner function. This is use
This is the Distributed Software (DSW) event device, which distributes
the task of scheduling events among all the eventdev ports and their
lcore threads.
DSW is primarily designed for atomic-only queues, but also supports
single-link and parallel queues.
(DSW would be more accurately described a
Signed-off-by: Mattias Rönnblom
---
config/common_base|5 +
drivers/event/Makefile|1 +
drivers/event/dsw/Makefile| 28 +
drivers/event/dsw/dsw_evdev.c | 361 +
drivers/event/dsw/dsw_evdev.h
I am seeing a build failure of MLX5 driver on Debian testing (buster)
with dpdk.org master branch.
This is using libibverbs-dev version 19.0-1
The failure is:
mlx5_glue.h:94:11: error: ‘struct ibv_counter_set_init_attr’ declared inside
parameter list will not be visible outside of this definitio
04/07/2018 17:22, Thomas Monjalon:
> 30/06/2018 01:10, Kevin Traynor:
> > Set the starting point that all commits on master branch
> > with Fixes tag should be backported to relevant stable/LTS
> > branches, and explain that the submitter may indicate it is
> > not suitable for backport.
> >
> > O
> -Original Message-
> From: Verma, Shally [mailto:shally.ve...@cavium.com]
> Sent: Wednesday, July 11, 2018 6:16 PM
> To: De Lara Guarch, Pablo
> Cc: dev@dpdk.org; Athreya, Narayana Prasad
> ; Challa, Mahipal
> ; Sahu, Sunila ;
> Gupta, Ashish
> Subject: RE: [PATCH v2 5/5] doc: add ZL
On Tue, 10 Jul 2018 09:59:58 -0700
Yipeng Wang wrote:
> +
> +static inline void
> +__hash_rw_reader_lock(const struct rte_hash *h)
> +{
> + if (h->readwrite_concur_support && h->hw_trans_mem_support)
> + rte_rwlock_read_lock_tm(h->readwrite_lock);
> + else if (h->readwrite_con
On Wed, 11 Jul 2018 20:53:46 +0300
Alex Kiselev wrote:
> librte_lpm: Improve lpm6 performance
>
> Rework the lpm6 rule subsystem and replace
> current rules algorithm complexity O(n)
> with hashtables which allow dealing with
> large (50k) rule sets.
>
> Signed-off-by: Alex Kiselev
> ---
> li
On Wed, Jul 11, 2018 at 09:22:33AM +0200, Nelio Laranjeiro wrote:
> Re-work flow engine to support port redirection actions through TC.
>
> This first series depends on [1] which is implemented in commit
> "net/mlx5: support inner RSS computation" and on [2].
> Next series will bring the port red
05/07/2018 17:47, Jasvinder Singh:
> Add connection agent to enable connectivity with external agen
> (e.g. telnet, netcat, Python script, etc).
>
> Signed-off-by: Cristian Dumitrescu
> Signed-off-by: Jasvinder Singh
> ---
> config/common_base | 2 +-
> config/
On Wed, Jul 11, 2018 at 09:22:47AM +0200, Nelio Laranjeiro wrote:
> Signed-off-by: Nelio Laranjeiro
> ---
> drivers/net/mlx5/mlx5_flow.c | 682 +++
> 1 file changed, 537 insertions(+), 145 deletions(-)
>
[...]
> @@ -1322,26 +1583,122 @@ mlx5_flow_merge(struct rte_
A quick reminder that the CFP for this closes on July 27th, which is two weeks
on Friday. If you're planning to submit a proposal you can do so at:
https://goo.gl/forms/Jh70MIA7QJE2nSBb2
Registration for the event is also open now. You can register at:
https://www.regonline.com/DPDKDublin
> --
29/06/2018 19:27, Gavin Hu:
> Gavin Hu (4):
> build: fix the meson build warning
> devtools: fix the missing ninja command error
> build: fix the meson cross compile error
> devtools: expand meson cross compiling test coverage
>
> gavin hu (2):
> mk: fix makefile based cross build errors
Hi Dmitri,
Thanks for debugging this.
However the below fix can truncate the data - I'll send an alternative fix
shortly.
Fiona
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Dmitry Eremin-Solenikov
> Sent: Wednesday, July 11, 2018 1:25 PM
> To: dev@dpdk.org
QAT driver checks byte alignment for KASUMI/SNOW 3G/ZUC algorithms using
cipher/auth_param, which are not initialized at this moment yet. Use
operation params instead.
Signed-off-by: Fiona Trahe
---
Change in v2 :
- fixed implementation to avoid data truncation
This fix should be backported to s
librte_lpm: Improve lpm6 performance
Rework the lpm6 rule subsystem and replace
current rules algorithm complexity O(n)
with hashtables which allow dealing with
large (50k) rule sets.
Signed-off-by: Alex Kiselev
---
lib/Makefile | 2 +-
lib/librte_lpm/meson.build | 1 +
lib/li
librte_lpm: Improve lpm6 performance
There is no need to rebuild the whole LPM tree
when a rule is deleted. This patch addresses this
issue introducing new delete operation.
Signed-off-by: Alex Kiselev
---
lib/librte_lpm/rte_lpm6.c | 698 ++
1 file ch
There is no need to rebuild the whole LPM tree
when a rule is deleted. This patch addresses this
issue introducing new delete operation.
Signed-off-by: Alex Kiselev
---
lib/librte_lpm/rte_lpm6.c | 698 ++
1 file changed, 583 insertions(+), 115 deletion
librte_lpm: Improve lpm6 performance
Rework the lpm6 rule subsystem and replace
current rules algorithm complexity O(n)
with hashtables which allow dealing with
large (50k) rule sets.
Signed-off-by: Alex Kiselev
---
lib/Makefile | 2 +-
lib/librte_lpm/meson.build | 1 +
lib/li
29/06/2018 19:27, Gavin Hu:
> The default test script covers only default host cc compiler, either gcc or
> clang, the fix is to increase the coverage by adding one more to cover
> clang and the others for gcc.
>
> Fixes: a55277a788 ("devtools: add test script for meson builds")
> Cc: sta...@dpdk.
05/07/2018 18:05, Ferruh Yigit:
> On 7/3/2018 5:16 PM, Emma Kenny wrote:
> > l2fwd_fork is not complied by default, this will make it compile
> >
> > Fixes: 95e8005a56e8 ("examples/l2fwd_fork: new app")
> >
> > Signed-off-by: Emma Kenny
>
> Acked-by: Ferruh Yigit
Applied, thanks
03/07/2018 14:10, Shahaf Shuler:
> Monday, July 2, 2018 6:40 PM, Emma Kenny:
> > Subject: [PATCH v1] examples/multi_process: fix build error
> >
> > Fix bug with undeclared variable name and calling a variable that is not
> > member of struct.
> >
> > CC main.o
> > l2fwd_fork/main.c: In function
>-Original Message-
>From: De Lara Guarch, Pablo [mailto:pablo.de.lara.gua...@intel.com]
>Sent: 11 July 2018 19:32
>To: Verma, Shally
>Cc: dev@dpdk.org; Athreya, Narayana Prasad
>; Challa, Mahipal
>; Sahu, Sunila ; Gupta,
>Ashish
>Subject: RE: [PATCH v2 5/5] doc: add ZLIB PMD documen
1 - 100 of 259 matches
Mail list logo
