The enic TSO implementation requires that the length of the Eth/IP/TCP
headers be passed to the NIC. Other than that, it's just a matter of
setting the mss and offload mode on a per packet basis.
In TSO mode, IP and TCP checksums are offloaded even if not requested
with mb->ol_flags.
Signed-off-b
Signed-off-by: Sameh Gobriel
Signed-off-by: Pablo de Lara
---
MAINTAINERS | 1 +
doc/guides/sample_app_ug/flow_distributor.rst | 492 ++
doc/guides/sample_app_ug/img/flow_distributor.svg | 417 ++
doc/guides/sample_a
This new sample app, based on the client/server sample app,
shows the user an scenario using the EFD library.
It consists of:
- A front-end server which has an EFD table that stores the
node id for each flow key, which will distribute the incoming
packets to the different nodes
- A back-end n
Elastic Flow Distributor (EFD) is a distributor library that uses
perfect hashing to determine a target/value for a given incoming flow key.
It has the following advantages:
- First, because it uses perfect hashing, it does not store
the key itself and hence lookup performance is not dependent
Signed-off-by: Byron Marohn
Signed-off-by: Karla Saur
Signed-off-by: Saikrishna Edupuganti
Signed-off-by: Pablo de Lara
---
MAINTAINERS | 1 +
app/test/Makefile| 3 +
app/test/test_efd.c | 494 +++
app/test/test_efd_perf
EFD is a distributor library that uses perfect hashing to determine a
target/value for a given incoming flow key. It has the following advantages:
first, because it uses perfect hashing it does not store the key itself and
hence lookup performance is not dependent on the key size. Second, the
targe
A referenced document in the Linux Kernel has been moved to a
sub-directory. And kernel community has moved to RST/Sphinx. The links are
replaced with HTML rendered links.
Signed-off-by: Yongseok Koh
---
doc/guides/contributing/patches.rst | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-
Add suupport for PCI IDs which was removed as a part of
commit 953158857f194991 ("net/bnxt: remove support for few devices")
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt_ethdev.c | 30 ++
1 file changed, 30 insertions(+)
diff --git a/drivers/net/bnxt/bnxt_e
On 12/30/2016 7:53 AM, Wei Zhao wrote:
> This patch adds a function to create the flow directory filter.
>
> Signed-off-by: Wei Zhao
> Signed-off-by: Wenzhuo Lu
>
> ---
>
> v2:
> --add new error set function
> ---
<...>
>
> +struct ixgbe_flow {
> + enum rte_filter_type filter_type;
> +
>From e346e359ed9c5e8261f09f93629bff56d7c10a11 Mon Sep 17 00:00:00 2001
From: "Jyoti, Anand B"
Date: Fri, 6 Jan 2017 08:40:55 +0530
Subject: [PATCH] examples/ip_pipeline: check vlan and mpls params
This commit add to CLI command check for the following errors
1. svlan and cvlan IDs greater than 1
On 12/30/2016 7:53 AM, Wei Zhao wrote:
> check if the rule is a TCP SYN rule, and get the SYN info.
>
> Signed-off-by: Wei Zhao
> Signed-off-by: Wenzhuo Lu
>
> ---
>
> v2:add new error set function
> ---
<...>
> +static int
> +ixgbe_parse_syn_filter(const struct rte_flow_attr *attr,
> +
On 12/30/2016 7:53 AM, Wei Zhao wrote:
> check if the rule is a ethertype rule, and get the ethertype info.
> Signed-off-by: Wei Zhao
> Signed-off-by: Wenzhuo Lu
>
> ---
>
> v2:add new error set function
> ---
<...>
> +static int
> +ixgbe_parse_ethertype_filter(const struct rte_flow_attr *att
On 01/04/2017 09:41 PM, Thomas Monjalon wrote:
2016-12-23 19:40, Tomasz Kulasek:
v15 changes:
- marked rte_eth_tx_prepare api as experimental
- improved doxygen comments for nb_seg_max and nb_mtu_seg_max fields
- removed unused "uint8_t tx_prepare" declaration from testpmd
No you didn't r
On 12/30/2016 7:53 AM, Wei Zhao wrote:
> Add rule validate function and check if the rule is a n-tuple rule,
> and get the n-tuple info.
>
> Signed-off-by: Wei Zhao
> Signed-off-by: Wenzhuo Lu
>
> ---
>
> v2:add new error set function
> ---
> drivers/net/ixgbe/ixgbe_ethdev.c | 414
>
On 12/30/2016 7:53 AM, Wei Zhao wrote:
> Add support for flush all the filters in SW.
>
> Signed-off-by: Wenzhuo Lu
> Signed-off-by: Wei Zhao
>
> ---
>
> v2:
> --change flush filter function call relationship
> ---
> drivers/net/ixgbe/ixgbe_ethdev.c | 118
> ++
On 12/30/2016 7:52 AM, Wei Zhao wrote:
> Add support for storing flow director filter in SW.
>
> Signed-off-by: Wenzhuo Lu
> Signed-off-by: Wei Zhao
> ---
>
> v2:
> --add a fdir initialization function in device start process
> ---
<...>
> @@ -1276,6 +1278,9 @@ eth_ixgbe_dev_init(struct rte_eth
On 12/30/2016 7:52 AM, Wei Zhao wrote:
> Add support for storing SYN filter in SW.
I think you have mentioned in previous review that you will update this
to "TCP SYN", I would like to remind in case it is missed.
>
> Signed-off-by: Wenzhuo Lu
> Signed-off-by: Wei Zhao
> ---
>
> v2:
> --synqf
On 12/22/2016 9:48 AM, Zhao1, Wei wrote:
> Hi, Yigit
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Wednesday, December 21, 2016 12:56 AM
>> To: Zhao1, Wei ; dev@dpdk.org
>> Cc: Lu, Wenzhuo
>> Subject: Re: [dpdk-dev] [PATCH 01/18] net/ixgbe: store SYN filter
>>
>> On 12/2/2016 10
diff --git a/examples/ip_pipeline/pipeline/pipeline_routing.c
b/examples/ip_pipeline/pipeline/pipeline_routing.c
index 3aadbf9..b7faf5b 100644
--- a/examples/ip_pipeline/pipeline/pipeline_routing.c
+++ b/examples/ip_pipeline/pipeline/pipeline_routing.c
@@ -494,6 +494,26 @@ app_pipeline_routing_add
On 05/01/2017 01:31, Yang, Qiming wrote:
[..]
+ if (strcmp(drvinfo->driver, "net_ixgbe") == 0)
Do you need this check. I think it is not good idea to add this kind
of checks into ethtool app. Why not just print "%d.%d.%d %#X, major,
minor, patch, etrack" for all cases ?
> Qiming: because
> -Original Message-
> From: Yongseok Koh [mailto:ys...@mellanox.com]
> Sent: Friday, January 6, 2017 1:23 AM
> To: Mcnamara, John
> Cc: dev@dpdk.org; Yongseok Koh
> Subject: [PATCH] doc: fix a link in contribution guide
>
> A referenced document in the Linux Kernel has been moved to a s
2017-01-06 17:14, Shreyansh Jain:
> On Wednesday 04 January 2017 03:35 AM, Thomas Monjalon wrote:
> > 2016-12-26 18:53, Shreyansh Jain:
> >> --- a/lib/librte_eal/common/include/rte_dev.h
> >> +++ b/lib/librte_eal/common/include/rte_dev.h
> >> @@ -152,6 +162,8 @@ struct rte_driver {
> >>struct r
2017-01-06 16:01, Shreyansh Jain:
> On Wednesday 04 January 2017 03:22 AM, Thomas Monjalon wrote:
> > 2016-12-26 18:53, Shreyansh Jain:
> >> +/**
> >> + * A structure describing a generic bus.
> >> + */
> >> +struct rte_bus {
> >> + TAILQ_ENTRY(rte_bus) next; /**< Next bus object in linked list
On Fri, Jan 06, 2017 at 02:12:48PM +0100, Thomas Monjalon wrote:
> 2017-01-06 18:16, Yuanhan Liu:
> > +static void
> > +eth_dev_init(struct rte_eth_dev *eth_dev, uint8_t port_id, const char
> > *name)
> > +{
> > + eth_dev->data = &rte_eth_dev_data[port_id];
> > + eth_dev->attached = DEV_ATTACH
On 1/6/2017 2:29 AM, Wu, Jingjing wrote:
>
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Wednesday, January 4, 2017 1:25 AM
>> To: Thomas Monjalon ; Ilya Maximets
>>
>> Cc: dev@dpdk.org; Zhang, Helin ; Ananyev, Konstantin
>> ; Wu, Jingjing ;
>> Heetae Ahn ; Richardson, Bruce
>>
On 1/4/2017 6:42 PM, Adrien Mazarguil wrote:
> Hi Ferruh,
>
> On Wed, Jan 04, 2017 at 05:49:46PM +, Ferruh Yigit wrote:
>> Hi Nelio,
>>
>> A quick question.
>
> I'll reply since it's related to the API.
>
>> On 12/29/2016 3:15 PM, Nelio Laranjeiro wrote:
>>> Introduce initial software for rt
2017-01-06 17:30, Shreyansh Jain:
> On Friday 06 January 2017 04:08 PM, Shreyansh Jain wrote:
> > On Wednesday 04 January 2017 03:16 AM, Thomas Monjalon wrote:
> >> 2016-12-26 18:53, Shreyansh Jain:
> >>> --- a/lib/librte_eal/linuxapp/eal/eal.c
> >>> +++ b/lib/librte_eal/linuxapp/eal/eal.c
> >>> @@
On 1/6/2017 8:50 AM, Adrien Mazarguil wrote:
> On Thu, Jan 05, 2017 at 04:49:31PM -0800, Yongseok Koh wrote:
>> Add PCI device ID for ConnectX-5 and enable multi-packet send for PF and VF
>> along with changing documentation and release note.
>>
>> Signed-off-by: Yongseok Koh
>
> Acked-by: Adrien
On 1/6/2017 8:16 AM, Rasesh Mody wrote:
> From: Harish Patil
>
> - Make qede_process_sg_pkts() inline and add unlikely check
> - Fix mbuf segment chaining logic in qede_process_sg_pkts()
> - Change qede_encode_sg_bd() to return total segments required
> - Fix first TX buffer descriptor's leng
2017-01-06 18:16, Yuanhan Liu:
> +static void
> +eth_dev_init(struct rte_eth_dev *eth_dev, uint8_t port_id, const char *name)
> +{
> + eth_dev->data = &rte_eth_dev_data[port_id];
> + eth_dev->attached = DEV_ATTACHED;
> + eth_dev_last_created_port = port_id;
> + nb_ports++;
> +
> +
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Slawomir
> Mrozowicz
> Sent: Thursday, January 05, 2017 4:50 PM
> To: dev@dpdk.org
> Cc: Mrozowicz, SlawomirX; Doherty, Declan; Kerlin, Marcin
> Subject: [dpdk-dev] [PATCH v2 2/3] lib/librte_cryptodev: functions f
On 1/6/2017 12:32 AM, Wu, Jingjing wrote:
>
>
>> -Original Message-
>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Wenzhuo Lu
>> Sent: Tuesday, January 3, 2017 2:55 PM
>> To: dev@dpdk.org
>> Cc: Yigit, Ferruh
>> Subject: [dpdk-dev] [PATCH v7 10/27] net/i40e: set VF MAC from PF s
The attached patch checks VLAN IDs and MPLS label for max value and also checks
the max number of supported MPLS labels to avoid array overflow in the CLI
command line parameters.
Regards,
Anand B Jyoti
On Wednesday 04 January 2017 03:43 AM, Thomas Monjalon wrote:
2016-12-26 18:54, Shreyansh Jain:
PCI scan and match now work on rte_device/rte_driver rather than PCI
specific objects. These functions can now be plugged to the generic
bus callbacks for scanning and matching devices/drivers.
Thes
Hi Jingjing,
> -Original Message-
> From: Wu, Jingjing
> Sent: Friday, January 6, 2017 1:29 AM
> To: Lu, Wenzhuo ; dev@dpdk.org
> Cc: Iremonger, Bernard ; sta...@dpdk.org
> Subject: RE: [dpdk-dev] [PATCH v7 26/27] net/i40e: fix segmentation fault in
> close
>
>
>
> > -Original Messa
On Friday 06 January 2017 04:08 PM, Shreyansh Jain wrote:
On Wednesday 04 January 2017 03:16 AM, Thomas Monjalon wrote:
2016-12-26 18:53, Shreyansh Jain:
--- a/lib/librte_eal/linuxapp/eal/eal.c
+++ b/lib/librte_eal/linuxapp/eal/eal.c
@@ -844,6 +845,9 @@ rte_eal_init(int argc, char **argv)
On 1/6/2017 5:27 AM, Beilei Xing wrote:
> The patch set depends on Adrien's Generic flow API(rte_flow).
>
> The patches mainly finish following functions:
> 1) Store and restore all kinds of filters.
> 2) Parse all kinds of filters.
> 3) Add flow validate function.
> 4) Add flow create function.
>
On Wednesday 04 January 2017 03:35 AM, Thomas Monjalon wrote:
2016-12-26 18:53, Shreyansh Jain:
--- a/lib/librte_eal/common/include/rte_dev.h
+++ b/lib/librte_eal/common/include/rte_dev.h
@@ -152,6 +162,8 @@ struct rte_driver {
struct rte_bus *bus; /**< Bus serviced by this dri
Hi Jinqjing,
> -Original Message-
> From: Wu, Jingjing
> Sent: Friday, January 6, 2017 12:37 AM
> To: Lu, Wenzhuo ; dev@dpdk.org
> Cc: Iremonger, Bernard
> Subject: RE: [dpdk-dev] [PATCH v7 17/27] net/i40e: set VF VLAN filter from
> PF
>
>
>
> > -Original Message-
> > From: dev
Hi Jinqjing, Wenzhuo,
> -Original Message-
> From: Lu, Wenzhuo
> Sent: Friday, January 6, 2017 8:20 AM
> To: Wu, Jingjing ; dev@dpdk.org
> Cc: Iremonger, Bernard
> Subject: RE: [dpdk-dev] [PATCH v7 14/27] net/i40e: set VF VLAN insertion
> from PF
>
> Hi Jingjing, Bernard,
>
>
> >
2017-01-05 14:22, Neil Horman:
> From: Neil Horman
>
> Coverity issue 139593 reports a forward null dereference from a for loop
> that works with a variable previously tested for null that had no error
> handling or condition to prevent it. Pretty obvious fix below.
>
> Signed-off-by: Neil Horm
> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Friday, December 23, 2016 8:25 AM
> To: dev@dpdk.org
> Cc: Trahe, Fiona; De Lara Guarch, Pablo; Griffin, John; Jain, Deepak K;
> Doherty, Declan; Kusztal, ArkadiuszX
> Subject: [PATCH v2 0/3] Fix iv sizes in crypto drivers capabilit
On Wednesday 04 January 2017 03:16 AM, Thomas Monjalon wrote:
2016-12-26 18:53, Shreyansh Jain:
--- a/lib/librte_eal/linuxapp/eal/eal.c
+++ b/lib/librte_eal/linuxapp/eal/eal.c
@@ -844,6 +845,9 @@ rte_eal_init(int argc, char **argv)
if (rte_eal_intr_init() < 0)
rte_panic("
CC'ing stable mailing list.
> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Friday, December 23, 2016 8:25 AM
> To: dev@dpdk.org
> Cc: Trahe, Fiona; De Lara Guarch, Pablo; Griffin, John; Jain, Deepak K;
> Doherty, Declan; Kusztal, ArkadiuszX
> Subject: [PATCH v2 2/3] crypto/aesni_
CC'ing stable mailing list.
> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Friday, December 23, 2016 8:25 AM
> To: dev@dpdk.org
> Cc: Trahe, Fiona; De Lara Guarch, Pablo; Griffin, John; Jain, Deepak K;
> Doherty, Declan; Kusztal, ArkadiuszX
> Subject: [PATCH v2 3/3] crypto/qat: f
On Wednesday 04 January 2017 03:22 AM, Thomas Monjalon wrote:
2016-12-26 18:53, Shreyansh Jain:
+DPDK_17.02 {
+ global:
+
+ rte_bus_list;
+ rte_eal_bus_add_device;
+ rte_eal_bus_add_driver;
+ rte_eal_bus_get;
+ rte_eal_bus_dump;
+ rte_eal_bus_register;
+
CC'ing stable mailing list.
> -Original Message-
> From: Kusztal, ArkadiuszX
> Sent: Friday, December 23, 2016 8:25 AM
> To: dev@dpdk.org
> Cc: Trahe, Fiona; De Lara Guarch, Pablo; Griffin, John; Jain, Deepak K;
> Doherty, Declan; Kusztal, ArkadiuszX
> Subject: [PATCH v2 1/3] crypto/aesni_
> -Original Message-
> From: Azarewicz, PiotrX T
> Sent: Monday, January 02, 2017 8:51 AM
> To: Azarewicz, PiotrX T; Kusztal, ArkadiuszX; dev@dpdk.org
> Cc: Trahe, Fiona; De Lara Guarch, Pablo; Griffin, John; Jain, Deepak K;
> Doherty, Declan; Kusztal, ArkadiuszX
> Subject: RE: [dpdk-dev]
Actually, virtio_hw->dev is not used since the beginning when it's
introduced. Remove it.
Signed-off-by: Yuanhan Liu
---
drivers/net/virtio/virtio_pci.c | 2 --
drivers/net/virtio/virtio_pci.h | 1 -
2 files changed, 3 deletions(-)
diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virt
The introduce of virtio 1.0 support brings yet another set of ops, badly,
it's not handled correctly, that it breaks the multiple process support.
The issue is the data/function pointer may vary from different processes,
and the old used to do one time set (for primary process only). That
said, th
We used to store the vtpci_ops at virtio_hw structure. The struct,
however, is stored in shared memory. That means only one value is
allowed. For the multiple process model, however, the address of
vtpci_ops should be different among different processes.
Take virtio PMD as example, the vtpci_ops i
Like vtpci_ops, the rte_pci_ioport has to store in local memory. This
is basically for the rte_pci_device field is allocated from process
local memory, but not from shared memory.
Signed-off-by: Yuanhan Liu
---
drivers/net/virtio/virtio_pci.c | 49 ++---
drive
If the primary enables the vector Rx/Tx path, the current code would
let the secondary always choose the non vector Rx/Tx path. This results
to a Rx/Tx method mismatch between primary and secondary process. Werid
errors then may happen, something like:
PMD: virtio_xmit_pkts() tx: virtqueue_enq
v3: - fixed several comments from Thomas regarding to eth_dev
- updated the release note and nic features matrix
v2: - fixed virtio 1.0 multiple process support
- fixed the case when few virtio net devices are managed by DPDK
while few others are handled by Linux kernel.
This patch
Assume we have two virtio ports, 00:03.0 and 00:04.0. The first one is
managed by the kernel driver, while the later one is managed by DPDK.
Now we start the primary process. 00:03.0 will be skipped by DPDK virtio
PMD driver (since it's being used by the kernel). 00:04.0 would be
successfully init
On 01/03/2017 03:20 PM, Ferruh Yigit wrote:
On 12/25/2016 10:33 PM, Thomas Monjalon wrote:
2016-12-23 16:57, Jan Blunck:
This repost addresses the review comments of Thomas Monjalon to completely
remove the ethdev helper to further decrease the coupling of the ethdev and
the eal layers. This re
Very good to know that!
Congratulations!
/Helin
-Original Message-
From: Martin Weiser [mailto:martin.wei...@allegro-packets.com]
Sent: Friday, January 6, 2017 5:17 PM
To: dev@dpdk.org; Ilya Maximets
Cc: Zhang, Helin ; Wu, Jingjing
Subject: Re: [dpdk-dev] XL710 with i40e driver drops p
Hello,
just to let you know we were finally able to resolve the issue. It seems
that the affected boards had a firmware issue with PCIe x8 v3.
When we forced the PCI slots to run at x8 v2 the issue disappeared for
Test 1 and Test 2. Test 3 still produced missed packets but probably due
to the redu
2017-01-06 09:50, Adrien Mazarguil:
> On Thu, Jan 05, 2017 at 04:49:31PM -0800, Yongseok Koh wrote:
> > Add PCI device ID for ConnectX-5 and enable multi-packet send for PF and VF
> > along with changing documentation and release note.
> >
> > Signed-off-by: Yongseok Koh
>
> Acked-by: Adrien Maz
Hi Jingjing,
> -Original Message-
> From: Wu, Jingjing
> Sent: Friday, January 6, 2017 8:33 AM
> To: Lu, Wenzhuo; dev@dpdk.org
> Cc: Lu, Wenzhuo
> Subject: RE: [dpdk-dev] [PATCH v7 03/27] net/i40e: set VF MAC anti-spoofing
> from PF
>
> > +
> > + vsi->info.valid_sections =
> > cpu_to_le
On Thu, Jan 05, 2017 at 04:49:31PM -0800, Yongseok Koh wrote:
> Add PCI device ID for ConnectX-5 and enable multi-packet send for PF and VF
> along with changing documentation and release note.
>
> Signed-off-by: Yongseok Koh
Acked-by: Adrien Mazarguil
Thanks!
--
Adrien Mazarguil
6WIND
From: Rasesh Mody
If value of number of rxq/txq is diffrent than
RTE_ETHDEV_QUEUE_STAT_CNTRS, limit per queue
stats/xstats to minimum of the two.
Fixes: 7634c5f91569 ("net/qede: add queue statistics")
Signed-off-by: Rasesh Mody
---
drivers/net/qede/qede_ethdev.c | 32 +++
From: Rasesh Mody
Signed-off-by: Rasesh Mody
---
drivers/net/qede/base/ecore_mcp.c |2 +-
drivers/net/qede/qede_ethdev.c| 11 +--
2 files changed, 6 insertions(+), 7 deletions(-)
diff --git a/drivers/net/qede/base/ecore_mcp.c
b/drivers/net/qede/base/ecore_mcp.c
index bb13828
Hi Jingjing, Bernard,
> -Original Message-
> From: Wu, Jingjing
> Sent: Friday, January 6, 2017 8:33 AM
> To: Lu, Wenzhuo; dev@dpdk.org
> Cc: Iremonger, Bernard
> Subject: RE: [dpdk-dev] [PATCH v7 14/27] net/i40e: set VF VLAN insertion from
> PF
>
>
>
> > -Original Message-
> >
From: Harish Patil
- Fix minimum RX buffer size to 1024B
- Force enable scatter/gather mode if given RX buf size is lesser than MTU
- Adjust RX buffer size to cache-line size with overhead included
Fixes: bec0228816c0 ("net/qede: support scatter gather")
Fixes: 2ea6f76aff40 ("qede: add core d
From: Harish Patil
Allocate double the number of fastpath status block index
since the PF RX/TX queues are not sharing the status block.
This is an interim solution till other parts of the code
is modified to handle the same.
Fixes: f1e4b6c0acee ("net/qede: fix status block index for VF queues")
From: Harish Patil
- Make qede_process_sg_pkts() inline and add unlikely check
- Fix mbuf segment chaining logic in qede_process_sg_pkts()
- Change qede_encode_sg_bd() to return total segments required
- Fix first TX buffer descriptor's length
- Replace repeatitive code using a macro
Fixes:
Hi Jingjing, Qi,
> -Original Message-
> From: Wu, Jingjing
> Sent: Friday, January 6, 2017 9:25 AM
> To: Lu, Wenzhuo; dev@dpdk.org
> Cc: Zhang, Qi Z
> Subject: RE: [dpdk-dev] [PATCH v7 25/27] net/i40e: set/clear VF stats from PF
>
>
>
> > -Original Message-
> > From: dev [mailt
68 matches
Mail list logo
