On Wed, 15 Mar 2017 04:45:03 +
Bruce Richardson wrote:
> On Tue, Mar 14, 2017 at 12:07:42PM +, Ferruh Yigit wrote:
> > On 3/14/2017 9:28 AM, raman geetha gopalakrishnan wrote:
> > > Hi ,
> > >
> > > Please find my query, Currently i am planning to develop DPDK APP (linux
> > > env). I
On Tue, Mar 14, 2017 at 10:31:40AM +0100, Gémes Géza wrote:
> 2017-03-14 10:28 keltezéssel, raman geetha gopalakrishnan írta:
> >Hi ,
> >
> >Please find my query, Currently i am planning to develop DPDK APP (linux
> >env). I do not have DPDK supported NIC. Till then i would still like to
> >develop
On Tue, Mar 14, 2017 at 12:07:42PM +, Ferruh Yigit wrote:
> On 3/14/2017 9:28 AM, raman geetha gopalakrishnan wrote:
> > Hi ,
> >
> > Please find my query, Currently i am planning to develop DPDK APP (linux
> > env). I do not have DPDK supported NIC. Till then i would still like to
> > develop
I've included a couple of specific comments inline below, and a general comment
here.
We have somebody proposing to add a new driver to DPDK. It's standalone and
doesn't affect any of the core libraries. They're willing to maintain the
driver and have included a patch to update the maintainers
On Tue, Mar 14, 2017 at 03:49:47PM +0100, Gaëtan Rivet wrote:
> > > The central question that I would like to tackle is this: why should
> > > we require from our users declaring a bonding device to have
> > > hot-plug support?
> > >
> > We'll, strictly speaking, I suppose we don't have to require
-Original Message-
From: Dai, Wei
Sent: Wednesday, March 15, 2017 9:39 AM
To: Wei, FangfangX
Subject: FW: [PATCH v2] examples/ip_fragmentation: fix check of packet type
> -Original Message-
> From: Dai, Wei
> Sent: Tuesday, March 14, 2017 10:48 PM
> To: dev@dpdk.org
> Cc: Anany
Hi John, Konstantin,
On Thu, Mar 9, 2017 at 4:07 PM, Ravi Kerur wrote:
> Hi John,
>
> On Thu, Mar 9, 2017 at 2:23 AM, Mcnamara, John
> wrote:
>
>> > From: Ravi Kerur [mailto:rke...@gmail.com]
>> > Sent: Monday, March 6, 2017 11:21 PM
>> > To: Mcnamara, John
>> > Cc: dev@dpdk.org; Ananyev, Kons
Hi Ferruh, Qi,
> > >> Subject: Re: [dpdk-dev] [PATCH 1/2] net/i40e: enable VF untag drop
> > >>
> > >> On 3/3/2017 1:59 AM, Qi Zhang wrote:
> > >>> Add a new private API to support the untag drop enable/disable for
> > >>> specific VF.
> > >>>
> > >>> Signed-off-by: Qi Zhang
> > >>> ---
> > >>>
On 3/14/2017 5:59 PM, Jan Mędala wrote:
<...>
>> In function ena_com_set_hash_ctrl(), the return value is assigned to
>> "ret" variable, but it is not returned. Fix it by adding the return.
>>
>> Signed-off-by: Yong Wang
> Acked-by: Jan Medala
Fixes: 99ecfbf845b3 ("ena: import communication
On 3/14/2017 5:38 PM, Ferruh Yigit wrote:
> On 3/1/2017 2:45 PM, Olivier Matz wrote:
>> It's not queue identifier but a descriptor identifier.
>>
>> Fixes: 4861cde46116 ("i40e: new poll mode driver")
>>
>> Signed-off-by: Olivier Matz
>
> Acked-by: Ferruh Yigit
Applied to dpdk-next-net/master, t
On 3/14/2017 5:47 PM, Ferruh Yigit wrote:
> On 3/13/2017 9:21 AM, Johan Samuelsson wrote:
>> Signed-off-by: Johan Samuelsson
>
> Acked-by: Ferruh Yigit
Fixes: 2269e7e815a2 ("pcap: add Rx and Tx byte counters")
Cc: sta...@dpdk.org
Applied to dpdk-next-net/master, thanks.
On 3/14/2017 4:38 PM, Yongseok Koh wrote:
>
>> On Mar 14, 2017, at 6:03 AM, Shahaf Shuler wrote:
>>
>> Prior to this patch only UIO/VFIO interrupt handlers types were supported.
>> This patch adds support for the external interrupt handler type, allowing
>> external drivers to set their own fds w
Acked-by: Jan Medala
Jan
2017-02-14 13:37 GMT+01:00 Yong Wang :
> Signed-off-by: Yong Wang
> ---
> drivers/net/ena/base/ena_com.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/ena/base/ena_com.c b/drivers/net/ena/base/ena_
> com.c
> index bd6f3c6..
Acked-by: Jan Medala
Jan
2017-02-14 13:37 GMT+01:00 Yong Wang :
> In function ena_com_set_hash_ctrl(), the return value is assigned to
> "ret" variable, but it is not returned. Fix it by adding the return.
>
> Signed-off-by: Yong Wang
> ---
> drivers/net/ena/base/ena_com.c | 2 +-
> 1 file
On 3/13/2017 9:21 AM, Johan Samuelsson wrote:
> Signed-off-by: Johan Samuelsson
Acked-by: Ferruh Yigit
On 3/1/2017 2:45 PM, Olivier Matz wrote:
> It's not queue identifier but a descriptor identifier.
>
> Fixes: 4861cde46116 ("i40e: new poll mode driver")
>
> Signed-off-by: Olivier Matz
Acked-by: Ferruh Yigit
Allain,
see inline,
+ I did restore the thread from
http://dpdk.org/ml/archives/dev/2017-March/060087.html into the same email.
To make it short, using ivshmem, you keep people unfocused from virtio.
Vincent,
Perhaps you can help me understand why the performance or functionality of
AVP vs.
This patch extend next_hop field from 8-bits to 21-bits in LPM library
for IPv6.
Added versioning symbols to functions and updated
library and applications that have a dependency on LPM library.
Signed-off-by: Vladyslav Buslov
Acked-by: Bruce Richardson
---
Fixed compilation error in l3fwd_lpm
On 3/14/2017 1:52 PM, Wiles, Keith wrote:
>
>> On Mar 14, 2017, at 8:51 PM, Adrien Mazarguil
>> wrote:
>>
>> Polling the Tx queue file descriptor before writing to it is not mandatory
>> since it is configured as non-blocking.
>>
>> Signed-off-by: Adrien Mazarguil
>> Acked-by: Pascal Mazon
>
On 3/14/2017 4:33 PM, Allain Legacy wrote:
> From: Matt Peters
>
> Allow the BAR setup to succeed if a device has at least 1 BAR region
> defined. Previously, the device probe would only succeed if at least one
> memory BAR existed, but there are devices that have only port I/O BARs.
>
> For ex
> On Mar 14, 2017, at 6:03 AM, Shahaf Shuler wrote:
>
> Implement rxq interrupt callbacks
>
> Signed-off-by: Shahaf Shuler
Acked-by: Yongseok Koh
Regards,
Yongseok
> On Mar 14, 2017, at 6:03 AM, Shahaf Shuler wrote:
>
> Prior to this patch only UIO/VFIO interrupt handlers types were supported.
> This patch adds support for the external interrupt handler type, allowing
> external drivers to set their own fds with specific interrupt handlers.
>
> Signed-off
On 03/14/2017 12:11 PM, Jan Blunck wrote:
On Mon, Mar 13, 2017 at 11:41 PM, Charles (Chas) Williams
wrote:
If the user reconfigures the queues size, then the previosly allocated
memzone may potentially be too small. Instead, always free the old
memzone and allocate a new one.
Fixes: dfaff37
From: Matt Peters
Allow the BAR setup to succeed if a device has at least 1 BAR region
defined. Previously, the device probe would only succeed if at least one
memory BAR existed, but there are devices that have only port I/O BARs.
For example, on Virtual Box a virtio device has only a single I
On Mon, Mar 13, 2017 at 11:41 PM, Charles (Chas) Williams
wrote:
> If the user reconfigures the queues size, then the previosly allocated
> memzone may potentially be too small. Instead, always free the old
> memzone and allocate a new one.
>
> Fixes: dfaff37fc46d ("vmxnet3: import new vmxnet3 po
On 3/13/2017 5:19 PM, Shrikrishna Khare wrote:
>
>
> On Mon, 13 Mar 2017, Andrew Rybchenko wrote:
>
>> Fixes: 4dd452523000 ("net/vmxnet3: add receive data ring support")
>>
>> Signed-off-by: Andrew Rybchenko
>
> Thank you for catching and fixing this.
>
> Acked-by: Shrikrishna Khare
Squashe
On 3/13/2017 8:02 AM, Jerin Jacob wrote:
> nicvf HW expects the DMA address of the packet data to be
> aligned with cache line size.
>
> Packet data offset is a function of struct mbuf size,
> mbuf private size and headroom. mbuf private size can
> be changed from the application in pool creation,
On 3/13/2017 3:15 AM, Lu, Wenzhuo wrote:
> Hi,
>
>> -Original Message-
>> From: Allain Legacy [mailto:allain.leg...@windriver.com]
>> Sent: Friday, March 10, 2017 10:38 PM
>> To: Lu, Wenzhuo
>> Cc: dev@dpdk.org
>> Subject: [PATCH] net/e1000: advertise offload capabilities for the EM PMD
>>
On 3/9/2017 4:21 PM, Andrew Rybchenko wrote:
> From: Ivan Malov
>
> According to 'libefx' API requirements, one is allowed to
> apply multicast address list to the port in started state
> only, otherwise the new array should be copied to a local
> storage in order to be applied during the next po
The rte_ipv4_fragment_packet API expects that the link/interface MTU value
passed in be divisible by 8 bytes. Given the name of the parameter is
"mtu" rather than "frag_size" it is not necessarily the case that it will
be divisible by 8. An MTU of 1500 happens to produce a max fragment size
of 14
2017-03-14 14:20, Raphael Cohn:
> Why it is surprising that I need to compile DPDK on a system Busybox?
>
> I currently build DPDK both for my local Alpine Linux system (busybox)
> using the system compiler, use busybox for my cross-tools toolchain, and
> also use it in Libertine Linux. It means a
The packet_type in mbuf is not correctly filled by ixgbe 82599 NIC.
To use the ether_type in ethernet header to check packet type is
more reliaber.
Fixes: 3c0184cc0c60 ("examples: replace some offload flags with packet type")
Fixes: ab351fe1c95c ("mbuf: remove packet type from offload flags")
Cc:
> The central question that I would like to tackle is this: why should
> we require from our users declaring a bonding device to have
> hot-plug support?
>
We'll, strictly speaking, I suppose we don't have to require it. But by that
same token, we don't need to do it in a separate PMD either, the
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
> Sent: Tuesday, March 14, 2017 2:31 AM
> To: Wu, Jingjing
> Cc: Zhang, Helin ; Chen, Jing D
> ; Yigit, Ferruh ;
> vincent.jar...@6wind.com; dev@dpdk.org
> Subject: Re: i40e SR-IOV with Linux PF
>
> 2017-03-
> -Original Message-
> From: Yigit, Ferruh
> Sent: Tuesday, March 14, 2017 9:32 PM
> To: Zhang, Qi Z ; Wu, Jingjing
> ; Zhang, Helin
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 2/2] app/testpmd: enable VF untag drop in
> testpmd
>
> On 3/9/2017 2:59 AM, Zhang, Qi Z wrote:
> >
>
> -Original Message-
> From: Yigit, Ferruh
> Sent: Tuesday, March 14, 2017 9:30 PM
> To: Zhang, Qi Z ; Wu, Jingjing
> ; Zhang, Helin
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 1/2] net/i40e: enable VF untag drop
>
> On 3/9/2017 3:24 AM, Zhang, Qi Z wrote:
> > Hi Ferruh:
> >
> >
Why it is surprising that I need to compile DPDK on a system Busybox?
I currently build DPDK both for my local Alpine Linux system (busybox)
using the system compiler, use busybox for my cross-tools toolchain, and
also use it in Libertine Linux. It means a system using DPDK can be much
more minima
> On Mar 14, 2017, at 4:29 PM, Pascal Mazon wrote:
>
> In the next patch, access to struct pmd_internals will be necessary in
> tap_flow.c to store the flows.
>
> Signed-off-by: Pascal Mazon
> Acked-by: Olga Shern
Acked-by: Keith Wiles
Regards,
Keith
> On Mar 14, 2017, at 4:29 PM, Pascal Mazon wrote:
>
> Each kernel netdevice may have queueing disciplines set for it, which
> determine how to handle the packet (mostly on egress). That's part of
> the TC (Traffic Control) mechanism.
>
> Through TC, it is possible to set filter rules that matc
> On Mar 14, 2017, at 4:34 PM, Pascal Mazon wrote:
>
> By default, a tap netdevice is of no use when not fed by a separate
> process. The ability to automatically feed it from another netdevice
> allows applications to capture any kind of traffic normally destined to
> the kernel stack.
>
> Thi
>
> +Cc dev@dpdk.org
>
> 2017-03-13 15:29, Thomas Monjalon:
> > Hi i40e developers,
> >
> > Referring to the VFD discussion, I thought basic behaviours were the
> > same regardless of the PF driver:
> > http://dpdk.org/ml/archives/dev/2016-December/053056.html
> > "
> > > In the meanwhile, we hav
> On Mar 14, 2017, at 8:51 PM, Adrien Mazarguil
> wrote:
>
> This commit adds a signal-based trigger to the Rx burst function in order
> to avoid unnecessary system calls while Rx queues are empty.
>
> Triggered Rx bursts put less pressure on the kernel, free up CPU resources
> for application
> On Mar 14, 2017, at 8:51 PM, Adrien Mazarguil
> wrote:
>
> Polling the Tx queue file descriptor before writing to it is not mandatory
> since it is configured as non-blocking.
>
> Signed-off-by: Adrien Mazarguil
> Acked-by: Pascal Mazon
Acked-by: Keith Wiles
> ---
> drivers/net/tap/rte_
On 3/9/2017 2:59 AM, Zhang, Qi Z wrote:
>
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Tuesday, March 7, 2017 7:14 PM
>> To: Zhang, Qi Z ; Wu, Jingjing ;
>> Zhang, Helin
>> Cc: dev@dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH 2/2] app/testpmd: enable VF untag drop in
>> testpmd
On 3/9/2017 3:24 AM, Zhang, Qi Z wrote:
> Hi Ferruh:
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Tuesday, March 7, 2017 6:51 PM
>> To: Zhang, Qi Z ; Wu, Jingjing ;
>> Zhang, Helin
>> Cc: dev@dpdk.org
>> Subject: Re: [dpdk-dev] [PATCH 1/2] net/i40e: enable VF untag drop
>>
>> O
Implement rxq interrupt callbacks
Signed-off-by: Shahaf Shuler
---
doc/guides/nics/features/mlx5.ini | 1 +
doc/guides/rel_notes/release_17_05.rst | 6 ++
drivers/net/mlx5/Makefile | 5 ++
drivers/net/mlx5/mlx5.c| 2 +
drivers/net/mlx5/mlx5_rxq.c
Prior to this patch only UIO/VFIO interrupt handlers types were supported.
This patch adds support for the external interrupt handler type, allowing
external drivers to set their own fds with specific interrupt handlers.
Signed-off-by: Shahaf Shuler
---
lib/librte_eal/linuxapp/eal/eal_interrupts
This patchset adds support for rxq interrupts on MLX5 PMD.
The first patch introduces changes on eal_interrupt to support external
interrupt handlers.
The second patch implements the support on mlx5 PMD and demonstrate the use of
the changes made in the first patch
on v2:
* add new function to
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Rename "rte_virtio_net.h" to "rte_vhost.h", to not let it be virtio
net specific.
Signed-off-by: Yuanhan Liu
---
doc/guides/rel_notes/deprecation.rst | 9 --
drivers/net/vhost/rte_eth_vhost.c | 2 +-
drivers/net/vhost/rte_eth_vhost.h
Polling the Tx queue file descriptor before writing to it is not mandatory
since it is configured as non-blocking.
Signed-off-by: Adrien Mazarguil
Acked-by: Pascal Mazon
---
drivers/net/tap/rte_eth_tap.c | 26 --
1 file changed, 8 insertions(+), 18 deletions(-)
diff --g
This commit adds a signal-based trigger to the Rx burst function in order
to avoid unnecessary system calls while Rx queues are empty.
Triggered Rx bursts put less pressure on the kernel, free up CPU resources
for applications and result in a noticeable performance improvement when
sharing CPU thr
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Rename "virtio-net" to "vhost" in the API comments.
Signed-off-by: Yuanhan Liu
---
lib/librte_vhost/rte_virtio_net.h | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
rename "virtio_net_device_ops" to "vhost_device_ops", to not let it
be virtio-net specific.
Signed-off-by: Yuanhan Liu
---
drivers/net/vhost/rte_eth_vhost.c | 2 +-
examples/tep_termination/main.c | 4 ++--
examples/vhost/main.c | 4 ++
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Include it internally, at vhost.h.
Signed-off-by: Yuanhan Liu
---
examples/vhost/main.h | 2 ++
lib/librte_vhost/rte_virtio_net.h | 4
lib/librte_vhost/vhost.h | 4
3 files changed, 6 insertions(+), 4 deletions(-)
Re
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
They are virti-net specific and should be defined inside the virtio-net
s/virti-net/virtio-net/
driver.
Signed-off-by: Yuanhan Liu
---
drivers/net/vhost/rte_eth_vhost.c | 2 ++
examples/tep_termination/main.h | 2 ++
examples/vhost/main.h
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Currently, we check vq->desc, vq->kickfd and vq->callfd to know whether
a virtio device is ready or not. However, we only do it when handling
SET_VRING_KICK message, which could be wrong if a vhost-user frontend
send SET_VRING_KICK first and SET_VRING_
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
We used to use rte_vhost_get_queue_num() for telling how many vrings.
However, the return value is the number of "queue pairs", which is
very virtio-net specific. To make it generic, we should return the
number of vrings instead, and let the driver do
On 3/12/2017 8:30 AM, Shahaf Shuler wrote:
> Thursday, March 9, 2017 8:10 PM, Thomas Monjalon:
>> 2017-02-08 15:57, Shahaf Shuler:
>>> /**
>>> + * It deletes registered eventfds.
>>> + *
>>> + * @param intr_handle
>>> + * Pointer to the interrupt handle.
>>> + */
>>> +void
>>> +rte_intr_free_epo
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
The queue pair is very virtio-net specific, other devices don't have
such concept. To make it generic, we should log the number of vrings
instead of the number of queue pairs.
This patch just does a simple convert, a later patch would export the
numbe
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Signed-off-by: Yuanhan Liu
---
lib/librte_vhost/rte_vhost_version.map | 1 +
lib/librte_vhost/rte_virtio_net.h | 21 +
lib/librte_vhost/vhost.h | 19 ---
lib/librte_vhost/virtio_net.c |
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Signed-off-by: Yuanhan Liu
---
lib/librte_vhost/rte_vhost_version.map | 1 +
lib/librte_vhost/rte_virtio_net.h | 13 +
lib/librte_vhost/vhost.c | 30 ++
lib/librte_vhost/vhost.h
On 3/14/2017 9:28 AM, raman geetha gopalakrishnan wrote:
> Hi ,
>
> Please find my query, Currently i am planning to develop DPDK APP (linux
> env). I do not have DPDK supported NIC. Till then i would still like to
> develop DPDK APP and want DPDK to use OS interface to TX/RX packets from
> NIC.
On 09/03/2017 04:48, Yerden Zhumabekov wrote:
Hello,
Can anybody explain why rte_malloc_heap.h is advertised as a public
interface? It only contains malloc_heap struct which is used
internally and violates DPDK naming conventions for public struct. I
haven't found any use of it outside of EAL
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Signed-off-by: Yuanhan Liu
---
lib/librte_vhost/rte_vhost_version.map | 1 +
lib/librte_vhost/rte_virtio_net.h | 1 +
lib/librte_vhost/vhost.c | 12
3 files changed, 14 insertions(+)
diff --git a/lib/librte_vhost/rt
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Some vhost-user driver may need this info to setup its own page tables
for GPA (guest physical addr) to HPA (host physical addr) translation.
SPDK (Storage Performance Development Kit) is one example.
Besides, by exporting this memory info, we could a
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Assume there is an application both support vhost-user net and
vhost-user scsi, the callback should be different. Making notify
ops per vhost driver allow application define different set of
callbacks for different driver.
Signed-off-by: Yuanhan Liu
On 10/3/2017 4:50 PM, Bruce Richardson wrote:
On Mon, Mar 06, 2017 at 09:10:30AM +, David Hunt wrote:
Signed-off-by: David Hunt
---
Please explain reason for change.
/Bruce
I've re-worked this change, as it's mostly do to with performance,
resulting in about 10% gain.
The number of Tx
On 10/3/2017 4:49 PM, Bruce Richardson wrote:
On Mon, Mar 06, 2017 at 09:10:29AM +, David Hunt wrote:
Signed-off-by: David Hunt
---
Title could do with some rewording - e.g. "make distributor API calls on
dedicated core"
This also requires an explanation as to why the change is being ma
On 10/3/2017 4:46 PM, Bruce Richardson wrote:
On Mon, Mar 06, 2017 at 09:10:27AM +, David Hunt wrote:
+ freq = rte_get_timer_hz();
+ t = rte_rdtsc() + freq;
+ while (!quit_signal_dist) {
+ if (t < rte_rdtsc()) {
+ print_stats();
+
On 10/3/2017 4:03 PM, Bruce Richardson wrote:
On Mon, Mar 06, 2017 at 09:10:19AM +, David Hunt wrote:
This patch includes public header file which will be used once
we add in the symbol versioning for v20 and v1705 APIs.
Also includes v1702 header file, and code for new
Now 1705.
burst
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Signed-off-by: Yuanhan Liu
---
examples/tep_termination/main.c| 4 +++-
examples/vhost/main.c | 43 +-
lib/librte_vhost/rte_vhost_version.map | 3 ---
lib/librte_vhost/rte_virtio_net.h |
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
vdev options is better for disabling/enabling some features.
Signed-off-by: Yuanhan Liu
---
drivers/net/vhost/rte_eth_vhost.c | 25
drivers/net/vhost/rte_eth_vhost.h | 30 -
dr
Thanks Monjalon For the quick reply.
On Tue, Mar 14, 2017 at 3:16 PM, Thomas Monjalon
wrote:
> 2017-03-14 14:58, raman geetha gopalakrishnan:
> > In what way i can still develop DPDK APP with non supported NIC till get
> > the DPDK NIC.
>
> You should check the bottom of this page (section Ohers
Thanks Keith For the quick Reply.
On Tue, Mar 14, 2017 at 3:22 PM, Wiles, Keith wrote:
>
>
> Sent from my iPhone
>
> > On Mar 14, 2017, at 5:29 PM, raman geetha gopalakrishnan <
> glowing...@gmail.com> wrote:
> >
> > Hi ,
> >
> > Please find my query, Currently i am planning to develop DPDK APP
virtio-user limits the qeueue number to 8 but provides no limit
check against the queue number input from user. If a bigger queue
number (> 8) is given, there is an overflow issue. Doing a sanity
check could avoid it.
Fixes: 37a7eb2ae816 ("net/virtio-user: add device emulation layer")
Cc: sta...@d
The first line of the MAINTAINERS file are:
"
The intention of this file is to provide a set of names that we can rely on
for helping in patch reviews and questions.
"
Unfortunately, some maintainers do not endorse their role for questions
asked on the mailing list.
Hope making it clear in the con
On 03/14/2017 10:46 AM, Maxime Coquelin wrote:
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Introduce few APIs to set/get/enable/disable driver features.
Signed-off-by: Yuanhan Liu
---
lib/librte_vhost/rte_vhost_version.map | 10
lib/librte_vhost/rte_virtio_net.h | 9
lib/li
Sent from my iPhone
> On Mar 14, 2017, at 5:29 PM, raman geetha gopalakrishnan
> wrote:
>
> Hi ,
>
> Please find my query, Currently i am planning to develop DPDK APP (linux
> env). I do not have DPDK supported NIC. Till then i would still like to
> develop DPDK APP and want DPDK to use OS
Hi Wei,
On Thu, 16 Feb 2017 17:49:22 +0100, Olivier Matz wrote:
> Hi Wei,
>
> On Thu, 9 Feb 2017 14:50:05 +, "Dai, Wei" wrote:
> > > -Original Message-
> > > From: Dai, Wei
> > > Sent: Thursday, February 9, 2017 10:38 PM
> > > To: 'Olivier Matz' ; dev@dpdk.org; Zhang,
> > > Helin ;
On 03/03/2017 10:51 AM, Yuanhan Liu wrote:
Introduce few APIs to set/get/enable/disable driver features.
Signed-off-by: Yuanhan Liu
---
lib/librte_vhost/rte_vhost_version.map | 10
lib/librte_vhost/rte_virtio_net.h | 9
lib/librte_vhost/socket.c | 90
2017-03-14 14:58, raman geetha gopalakrishnan:
> In what way i can still develop DPDK APP with non supported NIC till get
> the DPDK NIC.
You should check the bottom of this page (section Ohers):
http://dpdk.org/doc/nics
The simplest is to use the ring PMD which do not require a NIC at all
2017-03-14 07:58, Raphael Cohn:
> Hi,
>
> To compile DPDK on a system with Busybox tar installed, it's necessary to
> make a small change to the build system:-
>
> sed -i -e '/--keep-newer-files/d' mk/rte.sdkinstall.mk
> sed -i -e 's;--strip-components=1 \\;--strip-components=1;g' mk/
> r
On 10/3/2017 4:51 PM, Bruce Richardson wrote:
On Mon, Mar 06, 2017 at 09:10:31AM +, David Hunt wrote:
This so that with the increased amount of stats we are counting,
we don't interfere with the rx core.
Where are the stats being counted in the current code and how would they
interfere?
2017-03-14 10:28 keltezéssel, raman geetha gopalakrishnan írta:
Hi ,
Please find my query, Currently i am planning to develop DPDK APP (linux
env). I do not have DPDK supported NIC. Till then i would still like to
develop DPDK APP and want DPDK to use OS interface to TX/RX packets from
NIC. How
ping again
2017-03-08 12:25, Thomas Monjalon:
> ping - still waiting for answers to below questions, please
>
> 2017-02-16 15:55, Thomas Monjalon:
> > 2017-02-16 13:58, Wu, Jingjing:
> > > From: Thomas Monjalon
> > > >
> > > > Hi,
> > > >
> > > > When reading the documentation, it is not easy t
2017-03-14 04:44, Wu, Jingjing:
> From: Thomas Monjalon [mailto:thomas.monja...@6wind.com]
> > > Hi i40e developers,
> > >
> > > Referring to the VFD discussion, I thought basic behaviours were the
> > > same regardless of the PF driver:
> > > http://dpdk.org/ml/archives/dev/2016-December/053056.ht
Hi ,
Please find my query, Currently i am planning to develop DPDK APP (linux
env). I do not have DPDK supported NIC. Till then i would still like to
develop DPDK APP and want DPDK to use OS interface to TX/RX packets from
NIC. How can i make it? I went through KNI and my understanding is you
can
From: Hemant Agrawal
When possible, replace the uses of rte_mempool_create() with
the helper provided in librte_mbuf: rte_pktmbuf_pool_create().
This is the preferred way to create a mbuf pool.
This also updates the documentation.
Signed-off-by: Olivier Matz
Signed-off-by: Hemant Agrawal
Ack
On Tue, 7 Mar 2017 11:32:03 +, Bruce Richardson
wrote:
> NOTE: this set depends on the v2 cleanup set sent previously.
> http://dpdk.org/ml/archives/dev/2017-February/thread.html#58200
>
> This patchset make a set of, sometimes non-backward compatible, cleanup
> changes to the rte_rin
On Wed, 8 Mar 2017 12:08:42 +, Bruce Richardson
wrote:
> On Wed, Mar 08, 2017 at 11:22:40AM +0100, Olivier MATZ wrote:
> > On Tue, 7 Mar 2017 11:32:10 +, Bruce Richardson
> > wrote:
> > > The bulk fns for rings returns 0 for all elements enqueued and negative
> > > for no space. Chan
On Wed, 8 Mar 2017 12:06:54 +, Bruce Richardson
wrote:
> On Wed, Mar 08, 2017 at 11:49:06AM +0100, Olivier MATZ wrote:
> > On Thu, 23 Feb 2017 17:24:05 +, Bruce Richardson
> > wrote:
> > > We can write a single common function for head manipulation for enq
> > > and a common one for d
This patchset adds the special "remote" feature to the tap PMD, that
actually enables capturing traffic from another netdevice. This is
especially useful to get packets into the DPDK app, when the remote
netdevice has no DPDK support.
The "remote" feature requires flow API support as flow rules wi
By default, a tap netdevice is of no use when not fed by a separate
process. The ability to automatically feed it from another netdevice
allows applications to capture any kind of traffic normally destined to
the kernel stack.
This patch implements this ability through a new optional "remote"
para
Each kernel netdevice may have queueing disciplines set for it, which
determine how to handle the packet (mostly on egress). That's part of
the TC (Traffic Control) mechanism.
Through TC, it is possible to set filter rules that match specific
packets, and act according to what is in the rule. This
Supported flow rules are now mapped to TC rules on the tap netdevice.
The netlink message used for creating the TC rule is stored in struct
rte_flow. That way, by simply changing a metadata in it, we can require
for the rule deletion without further parsing.
Supported items:
- eth: src and dst (wi
The flow API provides the ability to classify packets received by a tap
netdevice.
This patch only implements skeleton functions for flow API support, no
patterns are supported yet.
Signed-off-by: Pascal Mazon
Acked-by: Olga Shern
---
doc/guides/nics/features/tap.ini | 1 +
drivers/net/tap/M
This series add support for the flow API in tap PMD.
It enables filtering specific packets incoming on the tap netdevice, to
process only desired ones. Under the hood, it uses kernel TC (traffic
control), which takes place very early in the stack, and supports most
common pattern items and actions
In the next patch, access to struct pmd_internals will be necessary in
tap_flow.c to store the flows.
Signed-off-by: Pascal Mazon
Acked-by: Olga Shern
---
drivers/net/tap/Makefile | 1 +
drivers/net/tap/rte_eth_tap.c | 35 ++--
drivers/net/tap/rte_eth_tap.h | 74 ++
On Tue, Mar 14, 2017 at 04:20:39PM +0800, Wenfeng Liu wrote:
> Hi Yuanhan,
>
> >On Mon, Mar 13, 2017 at 03:09:15PM +, Wenfeng Liu wrote:
> >> This commit fixes an array overflow when number of queues is higher than
> 8.
> >
> >Firstly, this commit log could be a bit more informative, to someth
A tap netdevice does not support flow control; ensure nothing but
RTE_FC_NONE mode can be set.
Signed-off-by: Pascal Mazon
---
doc/guides/nics/features/tap.ini | 1 +
drivers/net/tap/rte_eth_tap.c| 19 +++
2 files changed, 20 insertions(+)
diff --git a/doc/guides/nics/featu
1 - 100 of 115 matches
Mail list logo
