Add client and reconnect option to vhost pmd. reconnect only works when
client is given as well.
Cc: Tetsuya Mukawa
Signed-off-by: Yuanhan Liu
---
drivers/net/vhost/rte_eth_vhost.c | 54 ++-
1 file changed, 42 insertions(+), 12 deletions(-)
diff --git a/driv
Add --client and --reconnect option to enable the client mode and
reconnect mode, respectively. --rconnect works only when --client
is given as well.
Signed-off-by: Yuanhan Liu
---
examples/vhost/main.c | 23 +--
1 file changed, 21 insertions(+), 2 deletions(-)
diff --git a/
When DPDK app crashes (or quits, or gets killed), and when QEMU supports
reconnecting (patches have been sent, not merged yet), a restart of DPDK
app would get stale vring base from QEMU. That would break the kernel
virtio net completely, making it non-work any more, unless a driver reset
is done.
Allow reconnecting on failure when both RTE_VHOST_USER_RECONNECT and
RTE_VHOST_USER_CLIENT flags are set.
Reconnecting means two things here:
- when DPDK app starts first and QEMU (as the server) is not started,
without reconnecting, DPDK app would simply fail on vhost-user
registration.
- w
Add a new paramter (flags) to rte_vhost_driver_register(). DPDK
vhost-user acts as client mode when RTE_VHOST_USER_CLIENT flag
is set.
The flags would also allow future extensions without breaking the
API (again).
The rest is straingfoward then: allocate a unix socket, and
bind/listen for server,
DPDK vhost-user just acts as server so far, so, using a struct named
as "vhost_server" is okay. However, if we add the support of DPDK
vhost-user acting as client, it doesn't make sense any more. Here
rename it to "vhost_user_socket".
There was too much wrong for "connfd_ctx", but I think it's obv
Both the vhost-user backend (DPDK here) and frontend (QEMU) could be
server, as well as client. DPDK just acts as server so far. This patch
set would make it possible to act as both.
A new arg (flags) is introduced for API rte_vhost_driver_register(). And the
client mode is enabled when RTE_VHOST_
On Fri, 6 May 2016 21:05:02 +0100
Shen Wei wrote:
> --- a/lib/librte_hash/rte_cuckoo_hash.c
> +++ b/lib/librte_hash/rte_cuckoo_hash.c
> @@ -1,7 +1,7 @@
> /*-
> * BSD LICENSE
> *
> - * Copyright(c) 2010-2015 Intel Corporation. All rights reserved.
> + * Copyright(c) 2010-2016 Intel Cor
Hi,
I am doing experiments about packet classification algorithm and I found I
always get RX Error when the throughput is too high, so I did the following
experiments.
There are two PC servers (A and B), each of them has a Intel 82599ES 10G
with two ports(1 and 2). And they are connected to each
From: Harish Patil
Signed-off-by: Harish Patil
---
drivers/net/qede/qede_ethdev.h |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/qede/qede_ethdev.h b/drivers/net/qede/qede_ethdev.h
index b080f5f..abb33af 100644
--- a/drivers/net/qede/qede_ethdev.h
+++ b
From: Sony Chacko
This patch enables VF to VF traffic with unmatched destination address.
Signed-off-by: Sony Chacko
---
drivers/net/qede/base/ecore_l2.c | 19 +++-
drivers/net/qede/qede_eth_if.c | 45 +-
2 files changed, 39 insertions(+),
From: Harish Patil
Change details:
- Add device id to the PCI table
- Add polling for the slowpath events for CMT mode device
- Add prerequites to allow 100g mode
o Min number of queues needed is 2
o Only even number of queues are allowed
- Update documentation
Signed-off-by:
This patch adds support for extended statistics for QEDE PMD.
Signed-off-by: Rasesh Mody
---
doc/guides/nics/overview.rst |2 +-
drivers/net/qede/qede_ethdev.c | 152 +---
2 files changed, 145 insertions(+), 9 deletions(-)
diff --git a/doc/guides/nics/
From: Sony Chacko
Add support for mtu_set.
Signed-off-by: Sony Chacko
---
drivers/net/qede/qede_ethdev.c | 34 ++
1 file changed, 34 insertions(+)
diff --git a/drivers/net/qede/qede_ethdev.c b/drivers/net/qede/qede_ethdev.c
index 3754454..af16277 100644
--- a
From: Sony Chacko
Add support for reta_query.
Signed-off-by: Sony Chacko
---
drivers/net/qede/qede_ethdev.c | 27 +++
1 file changed, 27 insertions(+)
diff --git a/drivers/net/qede/qede_ethdev.c b/drivers/net/qede/qede_ethdev.c
index 2b91e09..3754454 100644
--- a/dri
From: Sony Chacko
Add support for reta_update.
Signed-off-by: Sony Chacko
---
doc/guides/nics/overview.rst |2 +-
drivers/net/qede/qede_ethdev.c | 36
2 files changed, 37 insertions(+), 1 deletion(-)
diff --git a/doc/guides/nics/overview.rst b/doc/
From: Sony Chacko
Add support for rss_hash_conf_get.
Signed-off-by: Sony Chacko
---
drivers/net/qede/qede_ethdev.c | 34 ++
1 file changed, 34 insertions(+)
diff --git a/drivers/net/qede/qede_ethdev.c b/drivers/net/qede/qede_ethdev.c
index 84ff6f8..6614632 10
From: Sony Chacko
Add support for rss_hash_update.
Signed-off-by: Sony Chacko
---
doc/guides/nics/overview.rst |2 +-
doc/guides/nics/qede.rst |3 +--
drivers/net/qede/qede_eth_if.h |1 +
drivers/net/qede/qede_ethdev.c | 43
driver
This patch set adds support for enabling 100G mode for QEDE PMD.
It also adds support for APIs like
- mtu_set
- reta_update
- reta_query
- rss_hash_update
- rss_hash_conf_get
- xstats_get
The changes include enablement of vf-vf traffic and updated the driver
version to 8.7.9.0_1.1.0.1
The pa
From: Harish Patil
Under certain scenarios, MFW periodically polls the driver
for LAN statistics. This patch implements the osal hook to
fill in the stats.
Fixes: ffa002d318d36 ("qede: add base driver")
Signed-off-by: Harish Patil
---
drivers/net/qede/base/bcm_osal.c | 21 ++
Rename RTE_LIBRTE_QEDE_DEBUG_DRV to RTE_LIBRTE_QEDE_DEBUG_DRIVER
Fixes: 425cba2a5176 ("qede: enable PMD build")
Fixes: 33e9ff1b72ca ("qede: add core driver")
Signed-off-by: Rasesh Mody
---
config/common_base |2 +-
doc/guides/nics/qede.rst |2 +-
drivers/net/qede/qede_logs
Introduced Intel TSX-enabled scalable multi-writer Cuckoo hash
insertion.
This patch introduced scalable multi-writer Cuckoo Hash insertion
based on a split Cuckoo Search and Move operation using Intel
TSX. It can do scalable hash insertion with 22 cores with little
performance loss and negligible
Instead of a hard-coded maximum receive length, allow the bond interface
to inherit this limit from the slave interfaces. This allows
an application that uses jumbo frames to pass realistic values to
rte_eth_dev_configure without causing an error.
Before the bond interface is configured
v2 changes:
- remove type cast on constant
- check max_rx_pktlen when adding a slave to make sure it is >= max
packet length of existing slave interfaces
v3 changes:
- allow slaves with any max rx packet length to be added to the bonding
interface until it is configured. After bond_ethde
From: Sankar Chokkalingam
The new command enables the execution of script-file for 'n' number of
times in regular intervals. It takes script-file, number of times to be
executed, interval between each execution as inputs.
Syntax: run
Usage: This command helps to collect statistics of ports a
From: Guruprasad Mukundarao
This script parses the application configuration file and detects all the
pipelines specified therein, and then, it generates all the possible mappings
of those pipelines on the specified CPU core-list.
As a result, each of the possible pipeline-to-core mappings is sa
bulk functions expect that all memory is set with zeros
Fixes: 67ebdbef0c31 ("examples/ip_pipeline: add bulk update of firewall rules")
Signed-off-by: Daniel Mrzyglod
---
examples/ip_pipeline/pipeline/pipeline_firewall_be.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git
From: Guruprasad Mukundarao
This script parses the application configuration file and detects all the
pipelines specified therein, and then, it generates all the possible mappings
of those pipelines on the specified CPU core-list.
As a result, each of the possible pipeline-to-core mappings is sa
>Hi,
>
>I am doing experiments about packet classification algorithm and I found I
>always get RX Error when the throughput is too high, so I did the following
>experiments.
>
>There are two PC servers (A and B), each of them has a Intel 82599ES 10G
>with two ports(1 and 2). And they are connected
>The format and layout of /proc/cpuinfo changes on different architectures,
>therefore the DPDK cpu_layout.py tool doesn't work.
>
>Could you please change the tool to use sysfs
>/sys/devices/system/cpu/cpuN/topology/
>instead? Then the script would be portable and use the same files as the DPDK
>
Hello Thomas,
On Fri, 06 May 2016 16:32:53 +0200
Thomas Monjalon wrote:
> It looks a lot too much tricky to be integrated without code comments ;)
> Please make a documentation effort.
I know it's not the brightly shining code... I focused mainly on the C API of
this. Thanks a lot for comments.
On Fri, 06 May 2016 16:01:08 +0200
Thomas Monjalon wrote:
> 2016-05-06 12:48, Jan Viktorin:
> > --- /dev/null
> > +++ b/app/test/resource.h
> > @@ -0,0 +1,61 @@
> > +/*-
> > + * BSD LICENSE
> [...]
> > + */
>
> Please include a multi-line comment here to explain what is a resource
> and wh
The LPM performance tests should not be mixed with
functional tests: they are long and show some statistics.
Signed-off-by: Thomas Monjalon
---
app/test/Makefile | 2 +
app/test/test_lpm.c | 211 +--
app/test/test_lpm6.c | 155 +---
All commands merged into one: cmd_action_parsed.
modified bulk command:
action flow bulk
File line format:
flow
meter 0 meter 1 meter 2
meter 3
policer 0policer 1
policer 2policer 3
port
at least one meter needs to be provided (can be any meter ID) and at
least one
This patch modifies flow classifications pipeline command line
interface. All commands are merged into one cmd_fc_parsed.
Additionally a classification for ipv6, ipv4 and qinq can be added from
configuration file.
1. flow add qinq bulk
File line format:
qinq port id
File line example:
qinq 1 2
Each command are merged into one: cmd_firewall_parsed.
ADD command format is changed:
p firewall add priority ipv4
port
and bulk command was modified:
1. firewall add bulk
File line format:
priority ipv4
port
(protomask is a hex value)
File line example:
priority 0 ipv4 1.
several routing commands are merged into two commands:
route and arp - these two commands are handled by cli library.
Rest of the commands are handled internaly by the pipeline code.
Signed-off-by: Piotr Azarewicz
---
examples/ip_pipeline/pipeline/pipeline_routing.c | 1774 ++
All commands are merged into one command:
cmd_link_parsed.
Signed-off-by: Tomasz Kulasek
---
examples/ip_pipeline/pipeline/pipeline_common_fe.c | 309 -
1 file changed, 111 insertions(+), 198 deletions(-)
diff --git a/examples/ip_pipeline/pipeline/pipeline_common_fe.c
b/exa
Add a couple of additional functions that will allow to parse many types
of input parameters, i.e.: bool, 16, 32, 64 bits, hex, etc.
Signed-off-by: Piotr Azarewicz
---
examples/ip_pipeline/Makefile | 1 +
examples/ip_pipeline/config_parse.c | 242 +
examples/ip_pipeline/parse
Using the latest librte_cmdline improvements, the CLI implementation
of the ip_pipeline application is streamlined and improved, which
results in eliminating thousands of lines of code from the application,
thus leading to code that is easier to maintain and extend.
Michal Jastrzebski (6):
examp
Hello,
On 09/ 1/15 07:34 AM, Zhang, Helin wrote:
>> -Original Message-
>> From: Vlad Zolotarov [mailto:vladz at cloudius-systems.com]
>> Sent: Monday, August 31, 2015 2:42 PM
>> To: dev at dpdk.org
>> Cc: Zhang, Helin; Ananyev, Konstantin; avi at cloudius-systems.com; Vlad
>> Zolotarov
>>
Signed-off-by: Sergio Gonzalez Monroy
---
doc/guides/sample_app_ug/ipsec_secgw.rst | 583 ---
1 file changed, 381 insertions(+), 202 deletions(-)
diff --git a/doc/guides/sample_app_ug/ipsec_secgw.rst
b/doc/guides/sample_app_ug/ipsec_secgw.rst
index c11c7e7..5dbee2e 1
IPSec transport mode support.
Signed-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/esp.c | 121 ++-
examples/ipsec-secgw/ipsec.h | 1 +
examples/ipsec-secgw/rt.c| 32
examples/ipsec-secgw/sa.c| 39 ++
exampl
Support IPSec IPv6 allowing IPv4/IPv6 traffic in IPv4 or IPv6 tunnel.
We need separate Routing (LPM) and SP (ACL) tables for IPv4 and IPv6,
but a common SA table.
Signed-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/Makefile | 5 +-
examples/ipsec-secgw/esp.c | 128
Modify the default SP config variables names to be consistent with SA.
The resulting naming convention is that variables with suffixes _out/_in
are the default for ep0 and the reverse for ep1.
Signed-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/sp.c | 14 +++---
1 file change
The application only checks that an SA shoudln't be NULL through ASSERT
(only when debugging is enabled) without properly dealing with the case
of not having an SA for the processed packet.
If no SA is found, drop the packet.
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample applicatio
Rework implementation moving from function pointers approach, where each
function implements very specific functionality, to a generic function
approach.
Signed-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/esp.c | 8 +-
examples/ipsec-secgw/esp.h | 9 +-
examples/ipsec-secgw/i
Add support for building the application with DEBUG=1.
This option adds the compiler stack protection flag and enables extra
output in the application.
Also remove unnecessary VPATH setup.
Signe-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/Makefile | 5 +++--
1 file changed, 3 insert
Building the application with -O3 and -fstack-protection (default in
Ubuntu) results in the following error:
*** stack smashing detected ***: ./build/ipsec-secgw terminated
The error is caused by storing an 8B value in a 4B variable.
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample a
Current code fails to correctly check padding sequence for inbound
packets.
Padding sequence starts on 1 but it checks for 0.
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample application")
Signed-off-by: Sergio Gonzalez Monroy
---
examples/ipsec-secgw/esp.c | 2 +-
1 file changed, 1
Update IPSec sample app with IPv6 and Transport mode support.
The series contains some bug fixes to facilitate patch merge.
Sergio Gonzalez Monroy (9):
examples/ipsec-secgw: fix esp padding check
examples/ipsec-secgw: fix stack smashing error
examples/ipsec-secgw: add build option and clean
Problem:
Now the i40e VF PMD driver does not support reset event from PF.
Customer want the user unaware VF reset feature in VF PMD driver,
with this feature, the application doesn't need to concern about
the VF reset procedure and all the application can work correctly
when VF is doing the reset.
Hi David,
On 5/6/2016 2:29 AM, David Hunt wrote:
> This patch set adds a fifo stack handler to the external mempool
> manager.
Just a minor confusion for me. Usually, we refer stack as LIFO and queue
as FIFO. So is there any particular reason why we call "stack (fifo)" here?
Thanks,
Jianfeng
It looks a lot too much tricky to be integrated without code comments ;)
Please make a documentation effort.
2016-05-06 12:48, Jan Viktorin:
> --- a/app/test/Makefile
> +++ b/app/test/Makefile
> @@ -33,6 +33,37 @@ include $(RTE_SDK)/mk/rte.vars.mk
>
> ifeq ($(CONFIG_RTE_APP_TEST),y)
A comment
On Thu, May 05, 2016 at 02:22:54PM +0800, Jingjing Wu wrote:
> Previously, if message is sent successfully, but no response is
> received, function "i40evf_execute_vf_cmd" will return without error.
> The root cause is value "err" is overwritten. This patch fixes it.
>
> Fixes: ae19955e7c86 ("i40e
On 5/6/2016 1:34 AM, Tan, Jianfeng wrote:
> Hi David,
>
>
> On 5/6/2016 2:29 AM, David Hunt wrote:
>> This patch set adds a fifo stack handler to the external mempool
>> manager.
>
> Just a minor confusion for me. Usually, we refer stack as LIFO and
> queue as FIFO. So is there any particular re
2016-05-06 12:48, Jan Viktorin:
> --- /dev/null
> +++ b/app/test/resource.h
> @@ -0,0 +1,61 @@
> +/*-
> + * BSD LICENSE
[...]
> + */
Please include a multi-line comment here to explain what is a resource
and why it is needed.
> +
> +#ifndef _RESOURCE_H_
> +#define _RESOURCE_H_
> +
|...]
|> +
>
On Wed, May 04, 2016 at 01:42:25PM +0800, Jingjing Wu wrote:
> Previously, for tunnel packets, such as VXLAN/NVGRE, the vlan
> tags of the inner header will be stripped without putting vlan
> info to descriptor, what is not expected behaviour.
> This patch fixes it by changing hardware configuratio
Signed-off-by: Jan Viktorin
---
lib/librte_ether/rte_ethdev.c | 127 +-
lib/librte_ether/rte_ethdev.h | 31 +++
2 files changed, 157 insertions(+), 1 deletion(-)
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index 4af2
Signed-off-by: Jan Viktorin
---
lib/librte_ether/rte_ethdev.c | 15 ++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index 9378a4a..4af2e5f 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/
We abstract access to the intr_handle here as we want to get
it either from the pci_dev or soc_dev.
Signed-off-by: Jan Viktorin
---
lib/librte_ether/rte_ethdev.c | 15 +--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/
Signed-off-by: Jan Viktorin
---
lib/librte_ether/rte_ethdev.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index 5474523..0b0dcbc 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/rte_ethdev.c
@@ -3077,6 +3077,
It is not necessary to place the rte_pci_driver at the beginning
of the rte_eth_dev struct anymore as we use the container_of macro
to get the parent pointer.
Signed-off-by: Jan Viktorin
---
lib/librte_ether/rte_ethdev.c | 4 ++--
lib/librte_ether/rte_ethdev.h | 2 +-
2 files changed, 3 insertio
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/include/rte_common.h | 16
1 file changed, 16 insertions(+)
diff --git a/lib/librte_eal/common/include/rte_common.h
b/lib/librte_eal/common/include/rte_common.h
index 332f2a4..a9b6792 100644
--- a/lib/librte_eal/common/inclu
The SoC devices can be sometimes DMA non-coherent. This means that we need
to take care of memory allocation for those. Generally, drivers assume that
every device is DMA coherent. If a driver supports a DMA non-coherent device,
it sets the RTE_SOC_DRV_ACCEPT_NONCC flag.
Note that the is_dma_coher
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/eal_common_soc.c | 6 ++
lib/librte_eal/common/eal_private.h| 10 ++
lib/librte_eal/linuxapp/eal/eal_soc.c | 11 +++
3 files changed, 27 insertions(+)
diff --git a/lib/librte_eal/common/eal_common_soc.c
b/lib/librte
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/eal_common_soc.c | 16 ++--
1 file changed, 10 insertions(+), 6 deletions(-)
diff --git a/lib/librte_eal/common/eal_common_soc.c
b/lib/librte_eal/common/eal_common_soc.c
index af4daa5..d178c48 100644
--- a/lib/librte_eal/common/e
The flags are copied from the PCI ones. They should be refactorized into a
general set of flags in the future.
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/include/rte_soc.h | 10 ++
1 file changed, 10 insertions(+)
diff --git a/lib/librte_eal/common/include/rte_soc.h
b/lib/li
No idea whether this is useful. Who creates the numa_node in the sysfs?
This can be probably dropped later.
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/eal_common_soc.c | 8
lib/librte_eal/common/include/rte_soc.h | 1 +
lib/librte_eal/linuxapp/eal/eal_soc.c | 26
The strict GCC rules do not allow to set the rte_soc_addr.compatible as it is
marked 'const'. However, we need it to be const because drivers will set it
statically by const strings. So this hack enables both.
Is there a better way to go?
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/in
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/include/rte_soc.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/lib/librte_eal/common/include/rte_soc.h
b/lib/librte_eal/common/include/rte_soc.h
index 3192121..4b4789b 100644
--- a/lib/librte_eal/common/include/rte_soc.h
+++ b/lib/li
Signed-off-by: Jan Viktorin
---
lib/librte_eal/bsdapp/eal/eal_soc.c | 12 ++
lib/librte_eal/bsdapp/eal/rte_eal_version.map | 2 +
lib/librte_eal/common/eal_common_soc.c | 13 +++
lib/librte_eal/common/include/rte_soc.h | 50 +
lib/li
We reuse the existing infrastructure (rte_eal_get_kernel_driver_by_path) here
however another possible implementation is by parsing the uevent file.
As there are no well-known driver for SoC infra, we does not detect any. This
will be changed in the future by checking for VFIO and UIO drivers.
Si
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/eal_common_soc.c | 5 +
lib/librte_eal/common/eal_private.h| 13 +
lib/librte_eal/linuxapp/eal/eal_soc.c | 11 +++
3 files changed, 29 insertions(+)
diff --git a/lib/librte_eal/common/eal_common_soc.c
b/lib/libr
This code is not tested. We assume to white/blacklist SoC devices by giving
the prefix "soc:" on the command line.
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/eal_common_dev.c | 31 ++-
lib/librte_eal/common/eal_common_devargs.c | 7 ++
lib/librte_eal/com
Signed-off-by: Jan Viktorin
---
lib/librte_eal/bsdapp/eal/rte_eal_version.map | 3 +
lib/librte_eal/common/eal_common_soc.c | 200
lib/librte_eal/common/include/rte_soc.h | 25 +++
lib/librte_eal/linuxapp/eal/eal.c | 4 +
lib/librte_e
Signed-off-by: Jan Viktorin
---
lib/librte_eal/bsdapp/eal/eal.c | 4
lib/librte_eal/common/eal_private.h | 10 ++
lib/librte_eal/linuxapp/eal/eal.c | 3 +++
lib/librte_eal/linuxapp/eal/eal_soc.c | 17 +
4 files changed, 34 insertions(+)
diff --git a/lib
This option has the same meaning for the SoC infra as the --no-pci
for the PCI infra.
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/eal_common_options.c | 5 +
lib/librte_eal/common/eal_internal_cfg.h | 1 +
lib/librte_eal/common/eal_options.h| 2 ++
3 files changed, 8 inse
Signed-off-by: Jan Viktorin
---
lib/librte_eal/bsdapp/eal/Makefile | 1 +
lib/librte_eal/bsdapp/eal/eal_soc.c | 40
lib/librte_eal/bsdapp/eal/rte_eal_version.map | 3 +
lib/librte_eal/common/eal_common_soc.c | 45
lib/librte_eal/common/include/r
Signed-off-by: Jan Viktorin
---
app/test/test_soc.c | 106
lib/librte_eal/bsdapp/eal/Makefile | 1 +
lib/librte_eal/bsdapp/eal/rte_eal_version.map | 3 +
lib/librte_eal/common/eal_common_soc.c | 55
lib
Define initial structures and functions for the SoC infrastructure. We support
only a very minimal functions now. More features will be added in the following
commits. It is to be refactored when a generic rte_device/driver pair is added
to DPDK.
Signed-off-by: Jan Viktorin
---
app/test/Makefile
The functions pci_map_resource, pci_unmap_resource are generic so the pci_
prefix can be omitted. The functions are moved to the eal_common_dev.c so
they can be reused by other infrastructure.
Signed-off-by: Jan Viktorin
---
lib/librte_eal/bsdapp/eal/eal_pci.c| 2 +-
lib/librte_eal/comm
Generalize the PCI-specific pci_get_kernel_driver_by_path. The function is
general enough, we have just moved it to eal.c, changed the prefix to rte_eal
and provided it privately to other parts of EAL.
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/eal_private.h | 14 ++
lib
Generalize the PCI-specific pci_unbind_kernel_driver. It is now divided into
two parts. First, determination of the path and string identification of the
device to be unbound. Second, the actual unbind operation which is generic.
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/eal_private.
The eal_parse_sysfs_value function accepts a filename however, such interface
introduces race-conditions to the code. Introduce the variant of this function
that accepts an already opened file instead of a filename.
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/eal_filesystem.h | 5
Signed-off-by: Jan Viktorin
---
lib/librte_eal/common/include/rte_dev.h | 8
lib/librte_eal/common/include/rte_pci.h | 10 +-
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/lib/librte_eal/common/include/rte_dev.h
b/lib/librte_eal/common/include/rte_dev.h
index b
Hello,
as a part of reworking the rte_(pci_)device/driver core of DPDK, I am
developing support for devices connected anotherway then by the PCI bus. I call
those devices _SoC devices_ and they are similar to the kernel's
platform_device. The goal is to have access to the on-chip MACs integrated
i
On Tue, May 03, 2016 at 10:38:21PM -0700, Rasesh Mody wrote:
> Enhance the stats_get() routine to display drop counters under
> imissed counter.
> Added extended stats get support to provide additional info.
> Encorporated review comment to rename some of the stats.
>
> Signed-off-by: Rasesh Mody
On Tue, May 03, 2016 at 10:38:20PM -0700, Rasesh Mody wrote:
> Change the Tx routine logic to ring the doorbell once per burst and not
> on every Tx packet. This driver-level optimization is necessary to achieve
> line rates for larger frame sizes (1k or more).
>
> Fixes: 540a211084a7 ("bnx2x: dri
On Tue, Apr 26, 2016 at 07:50:15PM -0700, John Daley wrote:
> Fixes: 7182d3e7d177 ("enic: expose Rx missed packets counter")
> Signed-off-by: John Daley
> ---
> drivers/net/enic/enic_main.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/enic/enic_main.c
dpdk_nic_bind will only handle Ethernet devices, but Mellanox ConnectX-3 Pro,
for example, is a Network class device, but not an Ethernet one. Even though
this allows other devices in the list, like Wireless devices, this should not be
a problem.
Signed-off-by: Thadeu Lima de Souza Cascardo
---
2016-05-06 13:51, Bruce Richardson:
> It takes the existing lpm large routing table information and converts it
> from a header file included at compile time to a resource linked in. This
> improves things in two ways:
> 1. Improves DPDK build time
> 2. Removes approx 1 million lines of code from o
Series-Acked-by: Deepak Kumar JAIN
-Original Message-
From: Kusztal, ArkadiuszX
Sent: Friday, May 6, 2016 11:22 AM
To: dev at dpdk.org
Cc: Trahe, Fiona ; Jain, Deepak K ; Griffin, John ; Kusztal, ArkadiuszX
Subject: [PATCH 0/2] Added AES counter mode capability
This patchset adds AES
With the new libsso library, buffers can be encrypted/decrypted,
providing an offset in bits, so an extra unit test has been
added to cover this case.
Signed-off-by: Pablo de Lara
---
app/test/test_cryptodev.c | 121 ++
1 file changed, 121 insertions(+
Snow3G PMD supports now buffers that are non byte multiple,
so tests to cover this case have been added.
Signed-off-by: Pablo de Lara
---
app/test/test_cryptodev.c | 50 ++
app/test/test_cryptodev_snow3g_hash_test_vectors.h | 107 +
2 files c
Signed-off-by: Pablo de Lara
---
app/test/test_cryptodev.c | 4
1 file changed, 4 insertions(+)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 1656ad0..6d0abf1 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -5181,6 +5181,10 @@ static stru
No padding was added in the input buffers for snow3G tests,
due to a wrong calculation of the length (should be multiple
of the block size). This fix takes into account the case
where the length is not byte multiple.
Fixes: 8bdf665fe6c0 ("app/test: add SNOW 3G")
Signed-off-by: Pablo de Lara
---
Instead of modifying the content of the buffers, to compare them
at bit-level, use the new macro TEST_ASSERT_BUFFERS_ARE_EQUAL_BIT,
which does not make any modifications in the buffers.
Signed-off-by: Pablo de Lara
---
app/test/test_cryptodev.c | 68 +-
Underlying libsso_snow3g library now supports bit-level
operations, so PMD has been updated to allow them.
Signed-off-by: Pablo de Lara
---
doc/guides/cryptodevs/snow3g.rst | 5 +-
drivers/crypto/snow3g/rte_snow3g_pmd.c | 129 ++---
2 files changed, 120 insert
With the library update, the way to compile the library
has changed, so documentation reflects this change.
Also, the patch to fix the compilation issues present with gcc > 5.0
has been removed, as the issues have been fixed in the library.
Signed-off-by: Pablo de Lara
---
doc/guides/cryptodevs/
1 - 100 of 230 matches
Mail list logo