Hi,
This is how I want my pipelines to work:-
Pipeline 4:
Has to get the source mac address and save it in 160th(160-167) meta data
filed in the mbuf.
Pipeline 5:
Will take the vlan id and store it in 168th(168-175) offset of the same mbuf
Pipeline 6:
Take the incoming port id which is in 23rd pos
Hi Jeff,
A few comments below:
On Sun, May 28, 2017 at 11:44:40PM +0800, Jeff Guo wrote:
For HW hotplug feature, we had already got upstream that removal event adding
from 6wind as bellow.
dependency of “add device removal event” patch set:
http://dpdk.org/dev/patchwork/patch/23693/
[dpdk-dev
remove/add blank line according to convensions
Shachar Beiser (1):
net/mlx5: implement drop action in hardware classifier
drivers/net/mlx5/Makefile| 5 +
drivers/net/mlx5/mlx5_flow.c | 17 +
2 files changed, 22 insertions(+)
--
1.8.3.1
On Mon, May 29, 2017 at 06:02:59PM -0700, Yongseok Koh wrote:
> A sanity check is required in priv_fdir_disable(). If resizing Rx queue
> fails, this can cause a crash by referencing a NULL pointer.
>
> Cc: sta...@dpdk.org
> Fixes: 76f5c99e6840 ("mlx5: support flow director")
> Fixes: 0cdddf4d0626
From: Michal Jastrzebski
This patch introduce a mechanism for running dpdk application with parameters
provided by configuration file. New API for cfgfile library allows to create
a cfgfile at runtime, add new section and add entry in a section - opens up
the possibility to have applications dyna
Extend existing cfgfile library with providing new API functions:
rte_cfgfile_create() - create new cfgfile object
rte_cfgfile_add_section() - add new section to existing cfgfile object
rte_cfgfile_add_entry() - add new entry to existing cfgfile object
in specified section
This modification allows
From: Kuba Kozak
added function rte_eal_configure which translate
options from config file into argc, **argv form.
changed function rte_eal_init to meld
argc, argv** options from config file with
these from command line and then parse as
before
Signed-off-by: Kuba Kozak
Suggested-by: Bruce Ric
From: Kuba Kozak
This patch shows how to pass arguments to application
using config.ini file.
If a --cfgfile-path option is passed into commandline
non EAL section, then the file is loaded and used by app.
If a file called config.ini is present in current working
directory, and no --cfgfile-pa
27/05/2017 05:46, Rahul Lakkireddy:
> This series of patches add support for Chelsio T6 family of adapters.
It is almost one year since your last contribution,
welcome back Rahul :)
During that time, rte_flow has been implemented in DPDK.
Do you plan to use it for Chelsio offloads, instead of flo
On 5/27/2017 4:46 AM, Rahul Lakkireddy wrote:
> Add code to detect and run T6 devices. Update PCI ID Device table
> with Chelsio T6 device ids and update documentation.
Can you please update DPDK supported NICs web page [1] too, with a
separate web patch [2] ?
[1] http://dpdk.org/doc/nics
[2] ht
On 5/27/2017 4:46 AM, Rahul Lakkireddy wrote:
> Add 25G and 100G link speeds and update supported port modules.
Not with this patch, but with separate patch, can it be possible to
update speed_capa in cxgbe_dev_info_get() to get device specific speed
capability, and update doc/guides/nics/features
On 5/27/2017 4:46 AM, Rahul Lakkireddy wrote:
> This series of patches add support for Chelsio T6 family of adapters.
>
> Patches 1 - 4 update CXGBE PMD to detect and run Chelsio T6 cards.
>
> Patches 5 - 6 add new link speeds and update port modules.
>
> Patch 7 updates information prints to ac
28/05/2017 17:40, Yongseok Koh:
> Vectorized Rx API for x86 is added.
>
> Signed-off-by: Yongseok Koh
> ---
> drivers/net/mlx5/Makefile | 1 +
> drivers/net/mlx5/mlx5_defs.h | 6 +
> drivers/net/mlx5/mlx5_ethdev.c | 8 +
> drivers/net/mlx5/mlx5_rxq.c| 39 +++-
> drivers/net/mlx
26/05/2017 18:52, Ferruh Yigit:
> We are looking for re-sending [1] the Kernel Control Path (KCP)
> with some updates [2].
>
> Mainly this is an usability improvement for DPDK.
>
> And a quick reminder about what KCP is:
>
> "KCP is Linux virtual network interface that can control DPDK ports".
>
On 5/27/2017 4:47 AM, Jingjing Wu wrote:
> i40e base code upate. The main changes are:
> - use virtchnl.h instead of i40e_virtchnl.h
> - add support for Adaptive Virtual Function
> - add new AQ commands for read/write PHY registers
> - add new phy types for 25G
Hi Jingjing,
Patchset giving bu
On 5/27/2017 4:47 AM, Rahul Lakkireddy wrote:
> This series of patches rework TX and RX path to reduce latency
> and improve performance.
>
> Patch 1 reduces latency for slow traffic by using status page update
> on RX path to process batch of packets and improves coalesce TX path
> to handle slow
HI Jiayu,
> -Original Message-
> From: Hu, Jiayu
> Sent: Tuesday, May 30, 2017 6:29 AM
> To: Ananyev, Konstantin
> Cc: dev@dpdk.org; Wiles, Keith ;
> yuanhan@linux.intel.com
> Subject: RE: [PATCH v3 1/3] lib: add Generic Receive Offload API framework
>
> Hi Konstantin,
>
> > -O
On 5/27/2017 8:55 AM, Andrew Rybchenko wrote:
> From: Andy Moreton
>
> MCDI results retuerned in req.emr_rc have already been translated
> from MC_CMD_ERR_* to errno names, so using an MC_CMD_ERR_* value
> is incorrect.
>
> Fixes: e7cd430c864f ("net/sfc/base: import SFN7xxx family support")
> Cc
On 5/27/2017 9:33 AM, Wei Dai wrote:
> This patch removes some some 1GBASE-T PHY access since the FW
> configures the PHY. SW shall not configure or initialize link.
> Accessing the PHY would require the use of MDI clause 22 which
> should be avoided in high layer driver code.
>
> Signed-off-by: W
On 5/28/2017 9:37 AM, Rasesh Mody wrote:
> From: Harish Patil
>
> Some PMDs need to know the tunnel type in order to handle advance TX
> features. This patch adds a new TX offload flag for MPLS-in-UDP packets.
>
> Signed-off-by: Harish Patil
> ---
> lib/librte_mbuf/rte_mbuf.c |2 ++
> lib/
On 5/29/2017 2:29 PM, Adrien Mazarguil wrote:
> On Mon, May 29, 2017 at 11:40:58AM +0200, Nelio Laranjeiro wrote:
>> Flow rules must be applied in the same order as they have been created and
>> thus destroyed in the reverse order.
>>
>> Fixes: 2097d0d1e2cc ("net/mlx5: support basic flow items and
On 5/29/2017 2:30 PM, Adrien Mazarguil wrote:
> On Mon, May 29, 2017 at 11:45:09AM +0200, Nelio Laranjeiro wrote:
>> Since commit 8f094a9ac5d7 ("mbuf: set mbuf fields while in pool"), some
>> fields are already initialised and do not need to be modified by the PMD
>> anymore.
>>
>> Signed-off-by: N
On 5/26/2017 2:18 PM, Charles (Chas) Williams wrote:
> When .mac_addr_add() was changed to allow a return code, ring was changed
> to return -ENOTSUP. This changes the behavior of the rte_ring driver in
> a way that does not maintain backward compatibility. Additionally, if
> the intent is to ret
Hi Zhang,
You should cram "flow API" somewhere in the title of such commits.
On Tue, May 23, 2017 at 07:28:54PM -0400, Qi Zhang wrote:
> Add new meta pattern item RTE_FLOW_TYPE_ITEM_ROUGHLY.
>
> This is for device that support no-perfect match option.
> Usually a no-perfect match is fast but the
On 5/30/2017 1:45 PM, Ferruh Yigit wrote:
> On 5/26/2017 2:18 PM, Charles (Chas) Williams wrote:
>> When .mac_addr_add() was changed to allow a return code, ring was changed
>> to return -ENOTSUP. This changes the behavior of the rte_ring driver in
>> a way that does not maintain backward compatib
Hi Ferruh,
> -Original Message-
> From: Yigit, Ferruh
> Sent: Thursday, May 25, 2017 4:47 PM
> To: dev@dpdk.org
> Cc: Yigit, Ferruh ; Mcnamara, John
> ; Tahhan, Maryam
> ; Iremonger, Bernard
> ; Adrien Mazarguil
>
> Subject: [RFC v3] flow_classify: add librte_flow_classify library
>
> Si
On 5/26/2017 2:50 PM, Andrew Rybchenko wrote:
> There is not point to check other Tx descriptors.
> It is important if Tx datapath does not reset Tx descriptor
> mbuf pointer to NULL on completion (EF10 simple Tx will do).
>
> Signed-off-by: Andrew Rybchenko
> Reviewed-by: Andy Moreton
> Reviewe
On Sun, May 28, 2017 at 07:11:38PM +0300, Moti Haimovsky wrote:
> Implement rxq interrupt callbacks
>
> Signed-off-by: Moti Haimovsky
Hi Moti,
While this patch is almost identical to its mlx5 counterpart, I have one
minor nit (see below). Besides that:
Acked-by: Adrien Mazarguil
> ---
> doc
Implement rxq interrupt callbacks
Based on implemantation done for mlx5
Signed-off-by: Moti Haimovsky
---
doc/guides/nics/features/mlx4.ini | 1 +
doc/guides/rel_notes/release_17_08.rst | 5 +
drivers/net/mlx4/mlx4.c| 218 -
drivers/net/ml
Hi Bruce,
> -Original Message-
> From: Richardson, Bruce
> Sent: Monday, May 29, 2017 8:19 PM
> To: Hu, Jiayu
> Cc: Ananyev, Konstantin ; dev@dpdk.org;
> Wiles, Keith ; yuanhan@linux.intel.com
> Subject: Re: [dpdk-dev] [PATCH v3 1/3] lib: add Generic Receive Offload API
> framework
>
A couple nit-picks on the test, but once resolved:
Acked-by: Gage Eads
> -Original Message-
> From: Van Haaren, Harry
> Sent: Monday, May 29, 2017 10:17 AM
> To: dev@dpdk.org
> Cc: jerin.ja...@caviumnetworks.com; Eads, Gage ;
> Van Haaren, Harry
> Subject: [PATCH] event/sw: fix cr
This patch set introduces an ethdev-based generic API for Traffic
Metering and Policing (MTR), which is yet another standard RX offload for
Ethernet devices.
Similar to rte_flow and rte_tm APIs, the configuration of MTR objects is
done in their own namespace (rte_mtr) within the librte_ether libra
Following similar approach to rte_flow and rte_tm for modularity reasons.
Signed-off-by: Cristian Dumitrescu
---
lib/librte_ether/rte_ethdev.c | 12
lib/librte_ether/rte_ethdev.h | 20
lib/librte_ether/rte_ether_version.map | 6 ++
3 files
Signed-off-by: Cristian Dumitrescu
---
MAINTAINERS| 4 +
lib/librte_ether/Makefile | 7 +-
lib/librte_ether/rte_ether_version.map | 8 +
lib/librte_ether/rte_mtr.c | 184 +
lib/librte_ether/rte_mtr.h | 465
Signed-off-by: Cristian Dumitrescu
---
lib/librte_ether/rte_flow.h | 22 ++
1 file changed, 22 insertions(+)
diff --git a/lib/librte_ether/rte_flow.h b/lib/librte_ether/rte_flow.h
index c47edbc..2942ca7 100644
--- a/lib/librte_ether/rte_flow.h
+++ b/lib/librte_ether/rte_flow.
No comment?
So the next dates will be
17.08
Proposal deadline: May 28
Integration deadline: June 29
Release: August 1
17.11
Proposal deadline: August 25, 2017
Integration deadline: September 29, 2017
Currently when a malloc_elem is split after resizing, any padding
present in the elem is ignored. This causes the resized elem to be too
small when padding is present, and user data can overwrite the beginning
of the following malloc_elem.
Solve this by including the size of the padding when comp
Currently when a malloc_elem is split after resizing, any padding
present in the elem is ignored. This causes the resized elem to be too
small when padding is present, and user data can overwrite the beginning
of the following malloc_elem.
Solve this by including the size of the padding when comp
On Mon, May 29, 2017 at 12:43 PM, Ferruh Yigit
wrote:
> > +int rte_pmd_bnxt_get_tx_drop_count(uint8_t port, uint64_t *count)
> > +{
> > + struct rte_eth_dev *dev;
> > + struct rte_eth_dev_info dev_info;
> > + struct bnxt *bp;
> > +
> > + dev = &rte_eth_devices[port];
> > + rte
If port LSC interrupt is configured, application will read link
status directly, so driver need to prepare that value in advance.
Fixes: 9ae6068c86da ("fm10k: add dev start/stop")
Cc: sta...@dpdk.org
Signed-off-by: Xiao Wang
---
drivers/net/fm10k/fm10k_ethdev.c | 5 +
1 file changed, 5 inse
This problem is caused by a missing set of E1000_RXCSUM_CRCOFL
in eth_igb_rx_init(), it should be set to enable SCTP packet
L4 checksum.If it is not set, the printf message in cksum fwd
about L4 SCTP cksum flag is error.for example, even if the sctp
packet is not do L4 checksum, the print message i
Hi, Ferruh
> -Original Message-
> From: Yigit, Ferruh
> Sent: Monday, May 29, 2017 7:01 PM
> To: Zhao1, Wei ; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 00/11] net/e1000: Consistent filter API
>
> On 5/23/2017 8:12 AM, Wei Zhao wrote:
> > The patches mainly finish following functions:
Hi Radu,
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Radu Nicolau
> Sent: Monday, May 29, 2017 11:23 PM
> To: dev@dpdk.org
> Cc: Yigit, Ferruh; Iremonger, Bernard; Nicolau, Radu
> Subject: [dpdk-dev] [PATCH v2] ethdev: moved bypass functions to ixgbe
> pmd
Hi,
> -Original Message-
> From: Zhao1, Wei
> Sent: Wednesday, May 31, 2017 11:03 AM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo; Zhao1, Wei
> Subject: [PATCH v3] net/e1000: fix checksum valid flags error
>
> This problem is caused by a missing set of E1000_RXCSUM_CRCOFL in
> eth_igb_rx_init(),
Hi Michal,
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Michal Jastrzebski
> Sent: Monday, May 22, 2017 10:32 PM
> To: dev@dpdk.org
> Cc: Pattan, Reshma ; Jain, Deepak K
> ; Van Haaren, Harry
> ; Jastrzebski, MichalX K
> ; Azarewicz, PiotrX T
>
> Subject: [d
Hi Cristian,
We thought of using compressed trie because in worst case it will become
the simple trie but in general cases if will avoid very costly memory
operation of accessing a new node.
On 29-05-2017 18:12, Dumitrescu, Cristian wrote:
-Original Message-
From: Richardson, Bruce
Se
46 matches
Mail list logo
