PKT_RX_FDIR_ID should be set only if flow_tag is neither non-zero nor
MLX5_FLOW_MARK_DEFAULT.
Fixes: 6cb559d67b83 ("net/mlx5: add vectorized Rx/Tx burst for x86")
Fixes: 570acdb1da8a ("net/mlx5: add vectorized Rx/Tx burst for ARM")
Cc: sta...@dpdk.org
Reported-by: Xueming Li
Signed-off-by: Yongs
As the dst_mac of allmulti is already masked with the mask, it has 0x01 in
the first octet. Checking the least significant bit only can't distinguish
it from broadcast or IPv6 multicast.
Fixes: bb47fb6e6067 ("net/mlx5: fix flow type for allmulti rules")
Cc: sta...@dpdk.org
Signed-off-by: Yongseok
-Original Message-
> Date: Tue, 9 Jan 2018 17:00:16 +
> From: "Eads, Gage"
> To: "dev@dpdk.org"
> CC: "jerin.ja...@caviumnetworks.com" ,
> "santosh.shu...@caviumnetworks.com" ,
> Hemant Agrawal , "nipun.gu...@nxp.com"
> , "Van Haaren, Harry" ,
> "Richardson, Bruce"
> Subject: Doc
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tonghao Zhang
> Sent: Friday, January 5, 2018 10:11 PM
> To: dev@dpdk.org
> Cc: Tonghao Zhang
> Subject: [dpdk-dev] [PATCH 2/5] net/ixgbevf: set the inter-interrupt interval
> for EITR.
>
> Set EITR interval as
On Wed, Jan 10, 2018 at 09:20:06PM +0100, Thomas Monjalon wrote:
> 08/01/2018 14:47, Pavan Nikhilesh:
> > Adding common test assertion macros for unit testing.
> > Taken from test/test.h.
> >
> > Signed-off-by: Pavan Nikhilesh
> > Acked-by: Jerin Jacob
> > ---
> > lib/librte_eal/common/Makefile
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tonghao Zhang
> Sent: Friday, January 5, 2018 10:11 PM
> To: dev@dpdk.org
> Cc: Tonghao Zhang
> Subject: [dpdk-dev] [PATCH 3/5] net/ixgbevf: save IXGBE_VTEIMS to intr-
> >mask for performance.
>
> If dpdk APPs c
> Date: Fri, 13 Oct 2017 18:32:18 +0200
>
> This patch adds the PCI IDs for Intel 82579LM.
>
>Signed-off-by: Markus Theil
Thanks for the patch.
Acked-by: Wenzhuo Lu
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tonghao Zhang
> Sent: Friday, January 5, 2018 10:11 PM
> To: dev@dpdk.org
> Cc: Tonghao Zhang
> Subject: [dpdk-dev] [PATCH 4/5] net/ixgbevf: add check for rte_intr_enable.
The patch is not only for ixgbevf, but
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Tonghao Zhang
> Sent: Friday, January 5, 2018 10:11 PM
> To: dev@dpdk.org
> Cc: Tonghao Zhang
> Subject: [dpdk-dev] [PATCH 1/5] net/ixgbevf: unregister irq handler when
> other interrupts not allowed.
>
> When w
Move the spinlock functions to the base code,
and make them inline.
This patch is for next-net and want to be squashed
to the commit in fixes tag.
Fixes: 8267939627e7 ("net/avf/base: add base code for avf PMD")
Signed-off-by: Wenzhuo Lu
---
drivers/net/avf/base/avf_osdep.h | 23 +
Remove the useless code.
This patch is for next-net and want to be squashed
to the commit in fixes tag.
Fixes: fcb27233824a ("net/avf: initialization of avf PMD")
Signed-off-by: Wenzhuo Lu
---
drivers/net/avf/avf.h | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/net/avf/avf.h b/dri
The optimization for AVF code.
As the AVF patches is in the next-net repo but not merged
to the master branch, proposed these patches that can be
merged to the previouse ones.
Sorry for the inconvenience.
Wenzhuo Lu (3):
net/avf/base: add spinlock functions
net/avf: remove spinlock functions
Move the spinlock functions to the base code,
and make them inline.
This patch is for next-net and want to be squashed
to the commit in fixes tag.
Fixes: fcb27233824a ("net/avf: initialization of avf PMD")
Signed-off-by: Wenzhuo Lu
---
drivers/net/avf/avf_ethdev.c | 24
I have also tried with multiple cpus using the command line-
./app/app/x86_64-native-linuxapp-gcc/pktgen -c 0x -n 4 -- -T -P -m
"[0:4].0,[5:8].1,[9-12].2,[13-16].3"
On Thursday 11 January 2018 10:20 AM, Shailja Pandey wrote:
The command line for Pktgen is-
./app/app/x86_64-native-linuxa
>On 1/5/2018 9:52 AM, Ravi Kumar wrote:
>> Signed-off-by: Ravi Kumar
>
>lgtm except below comments.
>
>At least share build error needs a new version of the patch.
>Can you able to complete licensing updates for next version?
>
Hi Ferruh,
Thanks a lot for reviewing our code. Good to know that it
> -Original Message-
>> From: Kumar, Ravi1 [mailto:ravi1.ku...@amd.com]
>> Sent: Wednesday, January 10, 2018 10:30 AM
>> To: De Lara Guarch, Pablo ;
>> dev@dpdk.org
>> Cc: Shippen, Greg
>> Subject: RE: [PATCH v3 01/19] crypto/ccp: add AMD ccp skeleton PMD
>>
>> >Hi Ravi,
>> >
>> >> -
> -Original Message-
> From: Xing, Beilei
> Sent: Tuesday, January 9, 2018 6:38 PM
> To: Zhang, Qi Z
> Cc: dev@dpdk.org; Wu, Jingjing ; sta...@dpdk.org
> Subject: [PATCH] net/i40e: fix setting MAC address on i40evf
>
> New mac address is copied to dev->data->mac_addrs[0] before calling
The command line for Pktgen is-
./app/app/x86_64-native-linuxapp-gcc/pktgen -c 0xfff -n 4 -- -T -P -m
"[2].0,[4].1,[8].2,[10].3"
Configuration of the machine is-
Architecture: x86_64
CPU op-mode(s):32-bit, 64-bit
Byte Order:Little Endian
CPU(s):32
O
If an invlaid number of RX or TX queues is configured from testpmd
command like "port config all rxq number" or "port config all txq number".
or from --rxq and --txq in the command to start testpmd. The global variable
nb_rxq or nb_txq is updated by the invalid input.
This can cause testpmd crash.
If an invalid number of RX queues is configured from testpmd run-time
command like "port config all rxq number" or from --rxq in the command
to start testpmd, the global variable nb_rxq is updated by this invalid
value without this patch. It may cause testpmd crash. This patch refuses
invalid rxq s
If an invalid number of TX queues is configured from testpmd run-time
command like "port config all txq number" or from --txq in the command
to start testpmd, the global variable nb_txq is updated by this invalid
value without this patch. It may cause testpmd crash. This patch refuses
invalid txq s
Hi Stephen,
> -Original Message-
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Thursday, January 11, 2018 1:18 AM
> To: Lu, Wenzhuo
> Cc: dev@dpdk.org; Wu, Jingjing
> Subject: Re: [dpdk-dev] [PATCH v6 02/14] net/avf: initialization of avf PMD
>
> On Wed, 10 Jan
We need the synchronous way for multi-process communication,
i.e., blockingly waiting for reply message when we send a request
to the peer process.
We add two APIs rte_eal_mp_request() and rte_eal_mp_reply() for
such use case. By invoking rte_eal_mp_request(), a request message
is sent out, and th
Previously, vfio uses its own private channel for the secondary
process to get container fd and group fd from the primary process.
This patch changes to use the generic mp channel.
Test:
1. Bind two NICs to vfio-pci.
2. Start the primary and secondary process.
$ (symmetric_mp) -c 2 -- -p
By the multi-process channel, we add an mp action named "proc".
As a secondary process starts, it sends a "proc add" message to
the primary.
As the primary finds a failure in sending message to a specific
secondary process, that secondary process is treated as exited;
and we remove it from the se
Previouly, there are three channels for multi-process
(i.e., primary/secondary) communication.
1. Config-file based channel, in which, the primary process writes
info into a pre-defined config file, and the secondary process
reads the info out.
2. vfio submodule has its own channel ba
v1->v2: (Address comments from Anatoly and Konstantin)
- Use datagram unix socket to supersede stream unix socket + epoll.
- Change the secondary add/del mechanism as now we use connection-less
channel.
- Add mp_mutex_action to sync action register/unregister/reference.
- Limit max length
I believe it's in the moderation queue because Sharmilla sent it from her
gmail address. Is there any chance you can look around in there and see if
you can free it? If not, I will resubmit with the Acked-by.
On Wed, Jan 10, 2018 at 3:42 PM, Ferruh Yigit
wrote:
> On 1/9/2018 3:32 PM, Doherty,
On Mon, Jan 08, 2018 at 02:37:15PM +, Remy Horton wrote:
> Port Representors provide a logical presentation in DPDK of VF (virtual
> function) ports for the purposes of control and monitoring. Each port
> representor device represents a single VF and is associated with it's
> parent physical
On 2018年01月10日 23:34, Ophir Munk wrote:
I agree that classifying directly through eBPF using socket filter is easier
than building a set of rules in TC, however there are more concerns that
justify using eBPF with TC.
1. An RSS rule can come in the middle of non-RSS rules. It seems natural to
Hi Stephen,
> -Original Message-
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Thursday, January 11, 2018 1:15 AM
> To: Lu, Wenzhuo
> Cc: dev@dpdk.org; Wu, Jingjing
> Subject: Re: [dpdk-dev] [PATCH v6 02/14] net/avf: initialization of avf PMD
>
> On Wed, 10 Jan 2
Hi,
Thanks for splitting the patches.
I will review the first one today. Please see below.
10/01/2018 10:12, Jeff Guo:
> --- /dev/null
> +++ b/lib/librte_eal/bsdapp/eal/eal_dev.c
> +int
> +rte_dev_monitor_start(void)
> +{
> + return -1;
> +}
> +
> +int
> +rte_dev_monitor_stop(void)
> +{
> +
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Wednesday, January 10, 2018 5:59 PM
> To: Dai, Wei ; Yang, Qiming ;
> Peng, Yuan ; Lu, Wenzhuo ;
> Wu, Jingjing
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [PATCH v3 0/2] app/testpmd: fix invalid rxq and txq nubmer
> setting
Signed-off-by: Erik Gabriel Carrillo
---
doc/api/doxy-api-index.md | 1 +
1 file changed, 1 insertion(+)
diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
index 3492702..3110658 100644
--- a/doc/api/doxy-api-index.md
+++ b/doc/api/doxy-api-index.md
@@ -46,6 +46,7 @@ The public A
Signed-off-by: Erik Gabriel Carrillo
---
doc/guides/rel_notes/release_18_02.rst | 6 ++
1 file changed, 6 insertions(+)
diff --git a/doc/guides/rel_notes/release_18_02.rst
b/doc/guides/rel_notes/release_18_02.rst
index 24b67bb..8eafcd3 100644
--- a/doc/guides/rel_notes/release_18_02.rst
+++
Signed-off-by: Erik Gabriel Carrillo
---
doc/guides/prog_guide/event_timer_adapter.rst | 301 ++
doc/guides/prog_guide/index.rst | 1 +
2 files changed, 302 insertions(+)
create mode 100644 doc/guides/prog_guide/event_timer_adapter.rst
diff --git a/doc/gu
Buffer timer expiry events generated while walking a "run list"
in rte_timer_manage, and burst enqueue them to an event device
to the extent possible.
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/rte_event_timer_adapter.c | 118 +++---
1 file changed, 108 inse
Signed-off-by: Erik Gabriel Carrillo
---
MAINTAINERS | 6 ++
1 file changed, 6 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index b51c2d0..55581c8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -291,6 +291,12 @@ F: lib/librte_eventdev/*eth_rx_adapter*
F: test/test/test_event_eth_rx_a
Add a test that creates an event timer and detects the generation of a
timer expiry event being scheduled through the software event device.
Signed-off-by: Erik Gabriel Carrillo
---
test/test/test_event_timer_adapter.c | 106 +++
1 file changed, 106 insertions(+)
Add support for the RTE_EVENT_TIMER_ADAPTER_F_SP_PUT flag, which indicates
that the API should be used in single-producer put mode.
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/rte_event_timer_adapter.c | 13 ++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --
Add a mode to the event timer adapter that allows the event timer arm
and cancel APIs to return immediately, rather than waiting for the service
to update the state of each timer.
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/rte_event_timer_adapter.c | 69 ++--
Add interfaces to support the collection, retrieval and resetting of
per-adapter statistics.
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/rte_event_timer_adapter.c | 79 +++
lib/librte_eventdev/rte_event_timer_adapter.h | 51 +++
lib/librte
Add a function that can be used to initialize event timers so that they
are in a known state before being used for arm or cancel operations.
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/rte_event_timer_adapter.c | 8
lib/librte_eventdev/rte_event_timer_adapter.h | 11 ++
Define the callback function for the service that corresponds to an
adapter instance, as well as the callback for expired timers that the
service manages.
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/rte_event_timer_adapter.c | 198 +-
lib/librte_eventdev/
Signed-off-by: Erik Gabriel Carrillo
---
lib/Makefile | 2 +-
lib/librte_eventdev/Makefile | 2 +-
lib/librte_eventdev/rte_event_timer_adapter.c | 155 +++---
mk/rte.app.mk | 2 +-
4 files
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/rte_event_timer_adapter.c | 11 +--
lib/librte_eventdev/rte_event_timer_adapter.h | 2 --
2 files changed, 1 insertion(+), 12 deletions(-)
diff --git a/lib/librte_eventdev/rte_event_timer_adapter.c
b/lib/librte_eventdev/rte_e
When the event timer adapter uses a software implementation, timer
management is performed on a service core using a service registered at
the time of adapter creation. This commit adds a helper function to get
the service id that can be used by an application to assign an lcore for
the service to
Signed-off-by: Erik Gabriel Carrillo
---
test/test/test_event_timer_adapter.c | 77 +++-
1 file changed, 76 insertions(+), 1 deletion(-)
diff --git a/test/test/test_event_timer_adapter.c
b/test/test/test_event_timer_adapter.c
index c8623aa..58cbaba 100644
--- a/t
Add a definition to the default software implementation for the entry
point that can fill out driver-specific information in the info struct.
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/rte_event_timer_adapter.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --
Exercise the create and free functions in the event timer adapter
auto-test.
Signed-off-by: Erik Gabriel Carrillo
---
test/test/test_event_timer_adapter.c | 88
1 file changed, 88 insertions(+)
diff --git a/test/test/test_event_timer_adapter.c
b/test/test/t
Add definitions to the default software implementation for the functions
that start and stop adapter instances.
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/rte_event_timer_adapter.c | 19 +++
1 file changed, 15 insertions(+), 4 deletions(-)
diff --git a/lib/libr
Add definitions for the functions that allocate and deallocate adapter
instances in the default software implementation.
Signed-off-by: Erik Gabriel Carrillo
---
lib/Makefile | 2 +-
lib/librte_eventdev/Makefile | 2 +-
lib/librte_eventdev/rt
Add initial infrastructure for event timer adapter auto-test.
Signed-off-by: Erik Gabriel Carrillo
---
test/test/Makefile | 1 +
test/test/test_event_timer_adapter.c | 177 +++
2 files changed, 178 insertions(+)
create mode 100644 test/test/te
This commit adds the logic that is shared by all event timer adapter
drivers; the common code handles instance allocation and some
initialization.
Signed-off-by: Erik Gabriel Carrillo
---
drivers/event/sw/sw_evdev.c | 18 +
lib/librte_eventdev/Makefile
If an eventdev PMD does not wish to provide event timer adapter ops
definitions, the library will fall back to a default software
implementation whose entry points are added by this commit.
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/rte_event_timer_adapter.c | 107 +
Signed-off-by: Erik Gabriel Carrillo
---
lib/librte_eventdev/Makefile | 1 +
lib/librte_eventdev/rte_event_timer_adapter.h | 566 ++
lib/librte_eventdev/rte_eventdev.h| 4 +-
3 files changed, 569 insertions(+), 2 deletions(-)
create mode 1
This patch set adds an initial implementation of the software driver
for the event timer adapter library, adds some unit tests, and adds
documentation. More testing is needed, but I'd like to get feedback on the
current state.
This patch set produces the following checkpatch warning:
"macro with
Hello everyone.
It's sad that my comments were unanswered.
I'm ok with the first two – they were mostly style-related.
But I made an investigation on the third one and it is a bug.
This is a description (from X710 datasheet) of flags sent to
mac_address_write command:
By bits:
0-7 – Reserved
8 –
Use the same value as in Linux driver.
Signed-off-by: Igor Ryzhov
---
drivers/net/i40e/i40e_ethdev.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 285d92b..93b9dd0 100644
--- a/drivers/net/i40e/i40e_eth
In some cases tx_nr_rings/rx_nr_rings may not be initialized.
In those scenarios, we may wrongly fail device initialization.
This patch fixes it. Use max_tx_rings/max_rx_rings instead.
Fixes: 2da0426e7cf8 ("net/bnxt: check on-chip resources")
Signed-off-by: Ajit Khaparde
---
drivers/net/bnxt/bnx
On 1/4/2018 4:01 PM, Thomas Monjalon wrote:
> This patchset aims to add notifications for ethdev ports
> added or removed.
> It will be especially useful for hotplug.
>
> v2 & v3 changes: gather more patches in the series
> v4 change: fix a variable wraparound as in Matan's v1
>
> Matan Azrad (3)
On 1/4/2018 4:01 PM, Thomas Monjalon wrote:
> From: Matan Azrad
>
> There are new Ethernet device events - NEW and DESTROY, and new option
> to register all ports by one call.
>
> Adjust application to aforementioned changes.
>
> Signed-off-by: Matan Azrad
Reviewed-by: Ferruh Yigit
On 1/5/2018 8:47 PM, Stephen Hemminger wrote:
> On Fri, 5 Jan 2018 18:38:55 +0100
> Thomas Monjalon wrote:
>
>> There are 3 kind of link data in ethdev:
>> - capabilities (rte_eth_dev_info)
>> - configuration (rte_eth_conf)
>> - status (rte_eth_link)
>>
>> A bit-field is used for
On 1/3/2018 10:32 AM, Olivier Matz wrote:
> Replace internal functions to get dma address them with their generic iova
> version. Also rename the 'data' variable to 'mbuf' for more readability.
>
> Signed-off-by: Olivier Matz
> Acked-by: Ajit Khaparde
Series applied to dpdk-next-net/master, tha
On 1/9/2018 3:32 PM, Doherty, Declan wrote:
> On 04/01/2018 6:34 AM, Sharmila Podury wrote:
>> From: Sharmila Podury
>>
>> Set the MTU for bonding device by calling .mtu_set for all
>> the slaves. Set the MTU only if all slaves support .mtu_set,
>> and there is no error returned from any slave.
>>
On 1/9/2018 11:34 AM, Declan Doherty wrote:
> change the xmit_hash functions to handle bursts of packet instead of
> single packets at a time, and update the affected tx_burst functions.
>
> Signed-off-by: Declan Doherty
> Signed-off-by: Keith Wiles
Applied to dpdk-next-net/master, thanks.
On 1/9/2018 4:15 PM, Doherty, Declan wrote:
> On 28/12/2017 2:12 AM, Chas Williams wrote:
>> From: Chas Williams
>>
>> Use rte_eth_dev_default_mac_addr_set() to change a slave MAC address.
>> mac_address_set() only updates the software copy and does nothing to
>> update the hardware.
>>
>> Signed-
On 1/8/2018 8:24 PM, Ajit Khaparde wrote:
> Please apply these patches.
>
> Ajit Khaparde (10):
> net/bnxt: check return values in bnxt_dev_init
> net/bnxt: fix double increment of idx during Tx ring alloc
> net/bnxt: parse checksum offload flags
> net/bnxt: fix grp_info usage
> net/bnxt
08/01/2018 14:47, Pavan Nikhilesh:
> Adding common test assertion macros for unit testing.
> Taken from test/test.h.
>
> Signed-off-by: Pavan Nikhilesh
> Acked-by: Jerin Jacob
> ---
> lib/librte_eal/common/Makefile | 2 +-
> lib/librte_eal/common/include/rte_test.h | 69
> ++
10/01/2018 20:16, Jerin Jacob:
> -Original Message-
> > Date: Fri, 15 Dec 2017 17:02:42 +0530
> > From: Pavan Nikhilesh Bhagavatula
> > To: "Ananyev, Konstantin" ,
> > "jerin.ja...@caviumnetworks.com" ,
> > "santosh.shu...@caviumnetworks.com" ,
> > "Richardson, Bruce" , "Van Haaren, Har
On Wed, Jan 10, 2018 at 04:53:53PM +, Van Haaren, Harry wrote:
> Replying to self...
>
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Van Haaren, Harry
> > Sent: Wednesday, January 10, 2018 4:45 PM
> > To: Pavan Nikhilesh ;
> > jerin.ja...@caviumnetworks
On 1/10/2018 10:46 AM, Hemant Agrawal wrote:
> This patch series add various improvement and performance related
> optimizations for DPAA PMD
>
> v3:
> - handling review comments from Ferruh
> - update the API doc for new PMD specific API
>
> v2:
> - fix the spelling of PORTALS
> - Add Akhil'
Hi Harry,
Thanks for the review.
On Wed, Jan 10, 2018 at 04:38:35PM +, Van Haaren, Harry wrote:
> > From: Pavan Nikhilesh [mailto:pbhagavat...@caviumnetworks.com]
> > Sent: Wednesday, January 10, 2018 2:52 PM
> > To: jerin.ja...@caviumnetworks.com; santosh.shu...@caviumnetworks.com; Van
> > H
On 1/8/2018 1:35 PM, Andrew Rybchenko wrote:
> Change license from BSD-2-Clause-FreeBSD to BSD-3-Clause.
> Bump copyright year.
>
> Signed-off-by: Andrew Rybchenko
Series applied to dpdk-next-net/master, thanks.
On 1/9/2018 8:24 PM, Andrew Rybchenko wrote:
> The minimum size of Rx/Tx descriptor rings supported by Solarflare HW is 512.
> It adds inconvenience to run DPDK applications which typically use smaller
> default values. Also smaller numbers of used descriptors is better for
> performance.
>
> The
10/01/2018 14:46, Doherty, Declan:
> On 09/01/2018 11:22 PM, Thomas Monjalon wrote:
> > Hi,
> >
> > 08/01/2018 15:37, Remy Horton:
> >> Port Representors provide a logical presentation in DPDK of VF (virtual
> >> function) ports for the purposes of control and monitoring. Each port
> >> represento
-Original Message-
> Date: Fri, 15 Dec 2017 17:02:42 +0530
> From: Pavan Nikhilesh Bhagavatula
> To: "Ananyev, Konstantin" ,
> "jerin.ja...@caviumnetworks.com" ,
> "santosh.shu...@caviumnetworks.com" ,
> "Richardson, Bruce" , "Van Haaren, Harry"
> , "Eads, Gage" ,
> "hemant.agra...@nx
On 1/10/2018 1:01 PM, Wenzhuo Lu wrote:
> Adaptive Virtual Function (AVF) Driver is VF driver which supports for all
> future Intel devices without requiring a VM update.
> It promises the basic high speed connectivity. And since this happens to be
> an adaptive VF driver, every new drop of the V
-Original Message-
> Date: Mon, 8 Jan 2018 19:06:01 +0530
> From: Jerin Jacob
> To: Pavan Nikhilesh
> CC: harry.van.haa...@intel.com, gage.e...@intel.com, dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] app/eventdev: fix event device queue count
> User-Agent: Mutt/1.9.2 (2017-12-15)
>
> -
-Original Message-
> Date: Wed, 10 Jan 2018 15:56:16 +0530
> From: Pavan Nikhilesh
> To: Gage Eads , jerin.ja...@caviumnetworks.com,
> harry.van.haa...@intel.com, santosh.shu...@caviumnetworks.com
> Cc: dev@dpdk.org
> Subject: Re: [PATCH] event/sw: remove stale IQ references when reconfig
-Original Message-
> Date: Wed, 10 Jan 2018 17:40:41 +0530
> From: Hemant Agrawal
> To: Jerin Jacob , Sunil Kumar Kori
>
> CC: "dev@dpdk.org"
> Subject: Re: [PATCH v2 12/12] doc: add DPAA eventdev guide
> User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101
> Thunder
-Original Message-
> Date: Wed, 10 Jan 2018 15:08:29 +
> From: "Liang, Ma"
> To: Jerin Jacob
> CC: dev@dpdk.org, harry.van.haa...@intel.com, bruce.richard...@intel.com,
> deepak.k.j...@intel.com, john.ge...@intel.com, peter.mccar...@intel.com,
> sea...@gmail.com, marko.kovace...@int
Adding the size of ibv_flow_spec_action_drop is missing. This can cause
overflow when converting rte_flow to ibv_flow and thus corrupt memory.
Fixes: 8086cf08b2f0 ("net/mlx5: handle RSS hash configuration in RSS flow")
Cc: sta...@dpdk.org
Signed-off-by: Yongseok Koh
Acked-by: Nelio Laranjeiro
-
Port Representors provide a logical presentation in DPDK of VF (virtual
function) ports for the purposes of control and monitoring. Each port
representor device represents a single VF and is associated with it's
parent physical function (PF) PMD which provides the back-end hooks for
the representor
Port Representors provide a logical presentation in DPDK of VF (virtual
function) ports for the purposes of control and monitoring. Each port
representor device represents a single VF and is associated with it's
parent physical function (PF) PMD which provides the back-end hooks for
the representor
Port Representors provide a logical presentation in DPDK of VF (virtual
function) ports for the purposes of control and monitoring. Each port
representor device represents a single VF and is associated with it's
parent physical function (PF) PMD which provides the back-end hooks for
the representor
Port Representors provide a logical presentation in DPDK of VF (virtual
function) ports for the purposes of control and monitoring. Each port
representor device represents a single VF and is associated with it's
parent physical function (PF) PMD which provides the back-end hooks for
the representor
Port Representors provide a logical presentation in DPDK of VF (virtual
function) ports for the purposes of control and monitoring. Each port
representor device represents a single VF and is associated with it's
parent physical function (PF) PMD which provides the back-end hooks for
the represe
Port Representors provide a logical presentation in DPDK of VF (virtual
function) ports for the purposes of control and monitoring. Each port
representor device represents a single VF and is associated with it's
parent physical function (PF) PMD which provides the back-end hooks for
the representor
If mlx5_dev_link_status_handler() is executed while canceling the alarm,
deadlock can happen because rte_eal_alarm_cancel() waits for all callbackes
to finish execution and both calls are protected by priv->lock.
Fixes: 198a3c339a8f ("mlx5: handle link status interrupts")
Cc: sta...@dpdk.org
Sign
On Wed, 10 Jan 2018 14:15:49 +0800
Wenzhuo Lu wrote:
> +/* spinlock func for base code */
> +void
> +avf_init_spinlock_d(struct avf_spinlock *sp)
> +{
> + rte_spinlock_init(&sp->spinlock);
> +}
> +
> +void
> +avf_acquire_spinlock_d(struct avf_spinlock *sp)
> +{
> + rte_spinlock_lock(&sp->
On Wed, 10 Jan 2018 14:15:49 +0800
Wenzhuo Lu wrote:
> +
> +#define AVF_MAX_NUM_QUEUES 16
> +/* Vlan table size */
> +#define AVF_VLAN_TB_SIZE (4096 / (CHAR_BIT * sizeof(uint32_t)))
You could use ETHER_MAX_VLAN_ID (which is 4095).
Also it is most efficient if bit tables use u
From: George Wilkie
With bonding, after sending sufficient ipv4 packets,
bond_ethdev_rx_burst_8023ad() no longer recognizes LACP packets
because the packet_type is set to RTE_PTYPE_L3_IPV4.
Ensure packet_type is reset for non-ipv4 packets in vmxnet3_rx_offload.
Signed-off-by: George Wilkie
---
Hi Konstantin
From: Ananyev, Konstantin, Wednesday, January 10, 2018 3:36 PM
> Hi Matan,
>
> Few comments from me below.
> BTW, do you plan to add ownership mandatory check in control path
> functions that change port configuration?
No.
> Konstantin
>
> > -Original Message-
> > From:
Replying to self...
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Van Haaren, Harry
> Sent: Wednesday, January 10, 2018 4:45 PM
> To: Pavan Nikhilesh ;
> jerin.ja...@caviumnetworks.com; santosh.shu...@caviumnetworks.com; Eads,
> Gage ; hemant.agra...@nxp.com;
> From: Pavan Nikhilesh [mailto:pbhagavat...@caviumnetworks.com]
> Sent: Wednesday, January 10, 2018 2:52 PM
> To: jerin.ja...@caviumnetworks.com; santosh.shu...@caviumnetworks.com; Van
> Haaren, Harry ; Eads, Gage
> ; hemant.agra...@nxp.com; nipun.gu...@nxp.com; Ma,
> Liang J
> Cc: dev@dpdk.org;
On Wed, 10 Jan 2018 15:07:14 +
Matan Azrad wrote:
> Hi Stephan
>
> From: Stephen Hemminger, Tuesday, January 9, 2018 8:51 PM
> > To: Matan Azrad
> > Cc: Ferruh Yigit ; Thomas Monjalon
> > ; dev@dpdk.org; Raslan Darawsheh
> >
> > Subject: Re: [PATCH v3 6/8] net/vdev_netvsc: skip routed netv
Acked-by: Pascal Mazon
On 10/01/2018 17:20, Moti Haimovsky wrote:
> Ethdev Rx offloads API has changed since:
> commit ce17eddefc20 ("ethdev: introduce Rx queue offloads API")
> This commit support the new Rx offloads API.
>
> Signed-off-by: Moti Haimovsky
> ---
> V4:
> Modifications according t
> From: Pavan Nikhilesh [mailto:pbhagavat...@caviumnetworks.com]
> Sent: Wednesday, January 10, 2018 2:52 PM
> To: jerin.ja...@caviumnetworks.com; santosh.shu...@caviumnetworks.com; Van
> Haaren, Harry ; Eads, Gage
> ; hemant.agra...@nxp.com; nipun.gu...@nxp.com; Ma,
> Liang J
> Cc: dev@dpdk.org;
On Wed, 10 Jan 2018 17:12:20 +0800
Jeff Guo wrote:
> +static int
> +dev_monitor_fd_new(void)
> +{
> +
> + int uevent_fd;
> +
> + uevent_fd = socket(PF_NETLINK, SOCK_RAW | SOCK_CLOEXEC |
> + SOCK_NONBLOCK,
> + NETLINK_KOBJECT_UEVENT);
> + if (uev
1 - 100 of 367 matches
Mail list logo
