On Sat, Apr 29, 2023 at 2:49 PM Cliff Burdick wrote:
> > So the only things we need are 2 functions, if I understand well:
> >
> > int rte_flow_to_text(const struct rte_flow*);
> > struct rte_flow *rte_flow_from_text(const char *);
> >
> > Here I assume the
>
>
> On Thu, 25 May 2023 05:36:02 +
> "Lombardo, Ed" wrote:
>
> > Hi,
> > I have two DPDK processes in our application, where one process
> allocates 1024 2MB hugepages and the second process allocates 8 1GB
> hugepages.
> > I am allocating hugepages in a script before the application
t size.
On Sat, Apr 29, 2023 at 2:40 PM Thomas Monjalon wrote:
> This thread is an API suggestion, it should be discussed in
> the developer mailing list (did the Cc here).
>
> 29/04/2023 16:23, Cliff Burdick:
> > > Would rather the flow parser was rewritten as well. Doing
ave spent on figuring on why a flow
spec didn't work properly, this could be a huge timesaver for new projects
like Tom mentioned.
On Fri, Apr 28, 2023 at 5:04 PM Stephen Hemminger <
step...@networkplumber.org> wrote:
> On Fri, 28 Apr 2023 12:13:26 -0700
> Cliff Burdick wrote:
>
constructing each flow by hand so that all these other
> applications can benefit and not be worries about bugs during conversions.
> >
> >
> >
> > On Thu, Apr 27, 2023, 01:37 Thomas Monjalon tho...@monjalon.net>> wrote:
> > 26/04/2023 07:47, David Marchand:
>
this as an API call
rather than a user constructing each flow by hand so that all these other
applications can benefit and not be worries about bugs during conversions.
On Thu, Apr 27, 2023, 01:37 Thomas Monjalon wrote:
> 26/04/2023 07:47, David Marchand:
> > On Wed, Apr 26, 2023 at 6:47 AM Clif
Does anyone know if a generic parser for flow strings exists anywhere? The
one inside of testpmd is ideal, but unfortunately it's self-contained and
not distributed as part of a normal DPDK install. This seems like something
that is likely reinvented over and over and it would be useful if there
gt;
> On Fri, Dec 2, 2022 at 5:50 PM Cliff Burdick wrote:
> >>
> >>
> >> Thank you for the information.
> >>
> >> I would need RDMA from a network card. Is there anything boiling in the
> pot?
> >>
> >> On Thu, Dec 1, 2022 at 7:
>
>
> Thank you for the information.
>
> I would need RDMA from a network card. Is there anything boiling in the
> pot?
>
> On Thu, Dec 1, 2022 at 7:24 PM Cliff Burdick wrote:
> >
> > If you are looking for RDMA to GPU devices, you can use gpudev in newer
>
If you are looking for RDMA to GPU devices, you can use gpudev in newer
DPDK versions to do it.
On Wed, Nov 30, 2022 at 2:02 AM Antonio Di Bacco
wrote:
> I would like to understand if DPDK will support officially RDMA
> through a set of APIs.
>
> Regards.
>
yperv_intf16384 1 mlx5_core
>>
>> Can you please help me in resolving this error.
>>
>> Thanks,
>> Animesh Tripathi
>>
>>
>> On Fri, Jul 8, 2022 at 11:42 AM animesh tripathi
>> wrote:
>>
>>> Hi Cliff,
>>>
Please paste the output of the meson command. Are you running as root?
On Thu, Jul 7, 2022, 06:13 animesh tripathi wrote:
> Hi Team,
>
> I have to run dpdk on mellanox nic. I have compiled dpdk 21.08 with the
> following commands:
>
> meson --prefix=$(pwd)/build build
> ninja -C build
> ninja
Hi Antonio, see this discussion:
http://mails.dpdk.org/archives/users/2021-November/006014.html
On Thu, Apr 14, 2022 at 12:57 PM Antonio Di Bacco
wrote:
> I imagine that there should be a way of sharing memory between two DPDK
> processes running in different docker containers. Probably the
-Original Message-
> From: Stephen Hemminger
> Sent: Tuesday, March 1, 2022 11:41 PM
> To: Cliff Burdick
> Cc: Lombardo, Ed ; users@dpdk.org
> Subject: Re: How to increase mbuf size in dpdk version 17.11
>
> External Email: This message originated outside of NETSCOUT. Do not cli
>
> Percpu: 2816 kB
>
> HardwareCorrupted: 0 kB
>
> AnonHugePages: 13228032 kB
>
> CmaTotal: 0 kB
>
> CmaFree: 0 kB
>
> HugePages_Total:1024
>
> HugePages_Free:0
>
> HugePages_Rsvd:0
>
&g
statistics available
>
> -----Original Message-
> From: Stephen Hemminger
> Sent: Tuesday, March 1, 2022 5:46 PM
> To: Cliff Burdick
> Cc: Lombardo, Ed ; users@dpdk.org
> Subject: Re: How to increase mbuf size in dpdk version 17.11
>
> External Email: This message orig
Can you verify how many buffers you're allocating? I don't see how many
you're allocating in this thread.
On Tue, Mar 1, 2022 at 1:30 PM Lombardo, Ed
wrote:
> Hi Stephen,
> The VM is configured to have 32 GB of memory.
> Will dpdk consume the 2GB of hugepage memory for the mbufs?
> I don't mind
One thing you can try is using the irqaffinity boot parameter to force
interrupts onto your non-isolated cores: For timer interrupts you can try
running in NO_HZ_FULL mode, but that may not work if you have other
userspace processes running on those cores. It might be worth confirming
that the
It should work for any card supporting GPUDirect. The requirement is you
would need nv_peer_mem running, and this is now included in the latest
NVIDIA drivers.
On Tue, Nov 23, 2021 at 7:06 AM sak...@fujitsu.com
wrote:
> Hello
>
> I have a question about (CUDA) DPDK supported GPU cards.
> If
> Mapping /var/run/dpdk into both containers as you said made it possible to
> use the multi-process feature in Docker containers.
>
> Regards
> Staffan
>
> Den tors 18 nov. 2021 kl 07:47 skrev Li Feng :
>
>> On Thu, Nov 18, 2021 at 2:23 PM Cliff Burdick wrote:
>>
I don't remember that being a problem. Are you starting them with two
different file prefixes and mounting the hugepages directory into both
containers?
On Wed, Nov 17, 2021, 22:20 Li Feng wrote:
> On Thu, Nov 18, 2021 at 2:11 PM Cliff Burdick wrote:
> >
> > Yes. Try mapping th
Yes. Try mapping the dpdk metadata directory (/var/run/dpdk) into both
containers from the host. You should be able to do the normal dual process
methods.
On Wed, Nov 17, 2021, 05:58 Staffan Wiklund wrote:
> Hello
>
> I wonder if it is possible to use the DPDK multi-process feature in Docker
>
There shouldn't be any problem doing that. Just pass a different port to
the sending functions.
On Thu, Aug 20, 2020, 03:51 Alireza Sanaee wrote:
> Dear all,
>
> I would like to send packet from multiple PMD port using one core, I'm
> just wondering if this is practical at all or not. Has
It should convert to network order, although many applications it won't
matter since they use all F's. If you follow the code in flow_filtering,
indeed it's using:
#define FULL_MASK 0x /* full mask */
So it won't make any difference. The example should probably be
updated, though..
On
Use "set txpkts" to set the packet size.
On Thu, Jul 9, 2020 at 2:25 AM Xu, Chenjie wrote:
> Hi all,
> Do you know how to specify the packet size when using testpmd to test the
> network performance?
>
> Best Regards,
> Xu, Chenjie
>
Can you paste the whole command you're running? --file-prefix with
unique strings afterward should allow you to do that.
On Sun, Jul 5, 2020 at 7:49 AM Yan Fridland wrote:
> Hi Guys,
>
> I am trying to use the --file-prefix to run several independent PRIMARY
> processes on the same VM
t; p_udp_hdr = (struct udp_hdr*)rte_pktmbuf_prepend(ptMbuf,
> >> (uint16_t)sizeof(struct udp_hdr));
> >> p_ip_hdr = (struct ipv4_hdr*)rte_pktmbuf_prepend(ptMbuf,
> >> (uint16_t)sizeof(struct ipv4_hdr));
> >>
> >> are you saying that those calls are wrong?
rather than guess).
>
> Thanks
>
>
> On Wed, Jul 1, 2020 at 3:59 PM Cliff Burdick wrote:
>
>> If you look at the code for rte_pktmbuf_prepend it appears to be just
>> incrementing data_len and pkt_len by the same amount. My guess is that
>> those fields were not
ytes) and then added the Ethernet and L3
> headers (42 bytes) by calling rte_pktmbuf_prepend().
>
> I guess only the first segment is getting transmitted?
>
>
> On Wed, Jul 1, 2020 at 1:57 PM Cliff Burdick wrote:
>
>> Are you setting data_len and packet_len in the mb
Sorry, I didn't see those below, but it looks incorrect. Those numbers
should match if you're using a single segment. My guess is the 42 is
truncating it to the size of the udp header part only.
On Wed, Jul 1, 2020, 06:04 David Aldrich
wrote:
> > Are you setting data_len and packet_len in the
Running it as a separate thread in your primary application will not slow
it down any more than running a second process, so I'd try that first.
Both primary and secondary processes have master lcores, which is where all
the initialization is done for pools and rings before spawning the packet
Are you setting data_len and packet_len in the mbuf before sending?
On Wed, Jul 1, 2020, 03:23 David Aldrich
wrote:
> Hi,
>
> I have a problem transmitting udp packets with dpdk-stable-18.11.8. I have
> posted a question about it on stackoverflow:
>
>
>
at the code to confirm, but also check what you're
setting this structures to.
On Sun, Jun 7, 2020, 10:11 Alex Kiselev wrote:
> On 2020-06-07 17:21, Cliff Burdick wrote:
> > The mbuf pool said be configured to be the size of the largest packet
> > you expect to receive. If you're
Kiselev wrote:
> On 2020-06-07 15:16, Cliff Burdick wrote:
> > That shouldn't matter. The mbuf size is allocated when you create the
> > mempool, and data_len/pkt_len are just to specify the size of the
> > total packet and each segment. The underlying storage size is still
>
, 2020, 02:59 Alex Kiselev wrote:
> On 2020-06-07 04:41, Cliff Burdick wrote:
> > I can't tell from your code, but you assigned nb_rx to the number of
> > packets received, but then used vec_size, which might be larger. Does
> > this happen if you use nb_rx in your loops?
>
&
I can't tell from your code, but you assigned nb_rx to the number of
packets received, but then used vec_size, which might be larger. Does this
happen if you use nb_rx in your loops?
On Sat, Jun 6, 2020 at 5:59 AM Alex Kiselev wrote:
>
>
> > 1 июня 2020 г., в 19:17, Stephen Hemminger
>
No, it doesn't. You can either have primary and secondary or multiple
primaries. Can you explain why B can't be your primary? If you just allow
it to configure all the nic pools that should solve your issue.
On Thu, Apr 2, 2020, 18:12 403975152 <403975...@qq.com> wrote:
> Hi,all,
>
>
> I am
Hi, I have a multicast flow at 239.0.12.1 that I want to send all
membership queries to that address (standard IGMP messages) to one queue,
and UDP port 555 on that IP to another queue.
I first tried to make a one rte_flow rule that matched a destination of
239.0.12.1 to port 0 with priority 1 go
If you're interested in this for GPUs, you should check out cuVNF here:
https://developer.nvidia.com/aerial-sdk
On Sun, Mar 1, 2020 at 3:33 AM Xueming(Steven) Li
wrote:
> With a quick hack on mlx5 pmd, it's possible to send RDMA operation with
> few changes. Performance result between 25Gb
Hi, I'm trying to figure out how to transmit a chained buffer, and I can't
find any examples or threads on this mailing list to do so. My assumptions
are:
1) Only the first mbuf contains the Ethernet and other packet headers, and
all the rest are essentially just appended to the end of that.
2)
Xiaolong, maybe it would be useful for every vendor to maintain a matrix of
every flow action/match that's supported and not for each PMD? I know we've
had this problem with Mellanox as well. It's likely only a handful of
vendors right now that support any form of rte_flow, so starting it soon
iff,
>
> Thank you for the info, it is really useful. I’ll search for the KB
> article.
>
> - Chintan
>
> On Jun 14, 2019, at 4:23 PM, Cliff Burdick wrote:
>
> Hi Chintan, yes, it works using those instructions. If you're using
> mellanox cards they have a KB artic
Hi, I'm seeing an issue with the ConnectX-5 where occasionally the
rte_eal_init cannot see some of the ports on the cards:
EAL: PCI device :bf:00.0 on NUMA socket 1
EAL: probe driver: 15b3:1017 net_mlx5
EAL: PCI device :bf:00.1 on NUMA socket 1
EAL: probe driver: 15b3:1017 net_mlx5
That's definitely interesting. Hopefully someone from mellanox can comment
on the performance impact since I haven't seen it qualified.
On Thu, Feb 28, 2019, 18:57 Arvind Narayanan wrote:
>
> On Thu, Feb 28, 2019, 8:23 PM Cliff Burdick wrote:
>
>> What size packets are you u
What type are flow are you creating? I believe #95 is simply that it's not
supported on that NIC. I'm not sure if an HP variant of the CX4 will
support different flows, but it should be traceable in the driver.
On Sat, Feb 16, 2019 at 6:20 AM Filip Janiszewski <
cont...@filipjaniszewski.com>
ioctl.
On Thu, Feb 14, 2019, 18:11 Stephen Hemminger On Thu, 14 Feb 2019 13:05:28 -0800
> Cliff Burdick wrote:
>
> > Hi, I'm trying to get DPDK working inside of a container deployed with
> > Kubernetes. It works great if I pass hostNetwork: true (effectively
> > net=
Hi, I'm trying to get DPDK working inside of a container deployed with
Kubernetes. It works great if I pass hostNetwork: true (effectively
net=host in Docker) to where the container sees all the host interfaces.
The problem with this is you lose all normal Kubernetes networking for
other non-DPDK
This problem happened to me, and the issue I had was you need to make sure
all mempools, whether generated with rte_mempool_create, or pktmbuf_create
are created in the primary process. Your secondary can then look up the
name of them. If you try to create one of these in the secondary process,
now it behaves like network stub. Just collects
> stats. Can I feed it to kni?
>
> On Tue, Dec 4, 2018 at 6:59 PM Cliff Burdick wrote:
>
>> Can you try zeroing out the offloads flag at the top?
>>
>> .rxmode = {
>> .mq_mode = ETH_MQ_RX_N
Can you try zeroing out the offloads flag at the top?
.rxmode = {
.mq_mode = ETH_MQ_RX_NONE,
.max_rx_pkt_len = ETHER_MAX_LEN,
.split_hdr_size = 0,
.offloads = 0,
},
On Tue, Dec 4, 2018 at 12:53 AM Alex Zelichenko
Answered my own question. For anyone looking for an example of how it
works, look at:
lib/librte_eal/common/malloc_mp.c
It appears to be used fairly extensively in there.
On Tue, Nov 20, 2018 at 11:35 AM Cliff Burdick wrote:
> Does anyone have an example of how to use the rte_mp_msg API
Does anyone have an example of how to use the rte_mp_msg API? I figured it
was as easy as just created an rte_mp_msg, and calling rte_mp_sendmsg().
That seems to give an error of "Unable to open directory", though.
, it
comes back reliably. For anyone else reading this, try moving your
rte_eal_init earlier in your startup to see if there's a particular
function causing it.
On Thu, Nov 15, 2018 at 12:33 PM Cliff Burdick wrote:
> Not sure how that is happening. 1GB hugepages are enabled on boot, and
>
, 2018 at 11:54 AM Kyle Larose wrote:
> On Thu, Nov 15, 2018 at 2:30 PM Cliff Burdick wrote:
> >
> > Hi, I'm updating to 18.11 from 17.03, and I noticed there a several new
> > warnings printing on startup in my primary process:
> >
> ...
> > EAL: WARNING! Ba
Hi, I'm updating to 18.11 from 17.03, and I noticed there a several new
warnings printing on startup in my primary process:
EAL: Detected 48 lcore(s)
EAL: Detected 2 NUMA nodes
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Probing VFIO support...
EAL: WARNING! Base virtual address
Hi, I'm trying to get a cooperative multi-process application working, but
running into an error that searching isn't helping on. I'm starting the
primary with the following parameters:
-c 3 -n 4 --proc-type=auto
and it's correctly recognized as primary:
EAL: Auto-detected process type: PRIMARY
Hi Ori, your patch has:
port_conf.txmode.offloads &= dev_info.rx_offload_capa;
Should that right side by rx_offload_capa instead?
On Mon, Nov 5, 2018 at 11:42 AM Cliff Burdick wrote:
> Thanks Ori. I confirmed that the latest patch works with VLAN tags as well.
>
> On Mon, Nov 5,
Thanks Ori. I confirmed that the latest patch works with VLAN tags as well.
On Mon, Nov 5, 2018 at 6:05 AM Ori Kam wrote:
> Just verified by running the code at least in Mellanox there is no
> difference.
>
>
>
> Best,
>
> Ori
>
>
>
> *From:* Cliff Burdick
s on the NIC but normally it should just be ETH / IPV4
> traffic.
>
>
>
> Best,
>
> Ori
>
>
>
> *From:* Cliff Burdick
> *Sent:* Monday, November 5, 2018 3:42 PM
> *To:* Ori Kam
> *Cc:* Shahaf Shuler ; users
> *Subject:* Re: [dpdk-users] flow_filtering failed
On Mon, Nov 5, 2018, 04:56 Kusuma DS Hii,
>
> I have some doubts regarding multiple process example application.
>
> I have 2 processes running in my setup using multiprocess client and server
> concept, am doing rx_burst in server process and tx_burst in client
> process. But the thing is i have
can look at them:
>
> http://patches.dpdk.org/patch/47806/
> http://patches.dpdk.org/patch/47807/
>
> Best,
> Ori
>
> > -Original Message-
> > From: Shahaf Shuler
> > Sent: Monday, November 5, 2018 9:10 AM
> > To: Cliff Burdick ; users ; Ori Kam
&g
Hi I'm trying to run the flow_filtering example on a connectx-5, and it
seems to fail with dpdk 18.08:
$ sudo ./build/flow -l 1 -n 1
EAL: Detected 48 lcore(s)
EAL: Detected 2 NUMA nodes
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Probing VFIO support...
EAL: PCI device :01:00.0
Have you tried allocating memory on both numa nodes to rule that out?
On Tue, Oct 30, 2018, 16:40 Raghu Gangi
wrote:
> Hi,
>
> I am currently facing issue in brining up DPDK application. It is failing
> with the following message. rte_errno is set to 12 in this scenario. (Out
> of memory).
>
>
Does anyone know if it's possible to do a hairpin mode using rte_flow? For
example, if I send a packet to 10.1.1.1, in tx queue 0, but I also have a
flow rule to send 10.1.1.1 to rx queue 1, is there a way to get the nic to
do that without sending the actual packet out? Mellanox appears to support
ok for is a sleep instruction executed by the adapter, which is
> buffered in order with the packets issued by rte_eth_tx_burst.
> (Plus some basic math rules how to convert packet sizes to durations,
> based on line speeds).
>
> Am Sa., 13. Okt. 2018 um 23:05 Uhr schrieb Cliff Bur
work.
>
>
>
> Kindest regards,
>
> Raslan Darawsheh
>
>
>
> *From:* Cliff Burdick
> *Sent:* Friday, October 12, 2018 7:11 PM
> *To:* Raslan Darawsheh
> *Cc:* katsikas...@gmail.com; users
> *Subject:* Re: [dpdk-users] Problem installing rules with counters on
Maybe I'm misunderstanding the problem, but do you need to transmit
anything? Can you just use the rte_cycles functions to sleep for the
remaining period in 20ms?
On Thu, Oct 11, 2018 at 2:04 AM Philipp B wrote:
> Hi all!
>
> I am working on an RTP test traffic generator. The basic idea is
>
Hi Raslan, can you confirm if rte_flow is partially broken in dpdk 18 on
mlx5? The example code for rte_flow is broken, and I didn't see any
responses here:
http://mails.dpdk.org/archives/users/2018-October/003510.html
On Thu, Oct 11, 2018 at 5:16 AM Raslan Darawsheh
wrote:
> Hi Georgios,
>
>
The easy workaround is to install the mellanox OFED package with the flags
--dpdk --upstream-libs.
On Thu, Oct 11, 2018 at 8:57 AM Anthony Hart wrote:
>
> Having problems compiling DPDK for the Mellanox PMD.
>
> For dpdk-18-08 I get...
>
> CC efx_phy.o
> In file included from
It was my mistake. It's a server haswell processor, but the BIOS had AES-NI
disabled. I enabled it, recompiled with CONFIG_RTE_MACHINE="haswell", and
it works great now. Thanks!
On Tue, Oct 9, 2018 at 1:00 PM Wiles, Keith wrote:
>
>
> > On Oct 9, 2018, at 9:56 AM
he gcc manual specifying some other AES feature that's not what DPDK
> is listing?
>
>
>
> On Tue, Oct 9, 2018 at 6:54 AM Wiles, Keith wrote:
>
>>
>>
>> > On Oct 8, 2018, at 11:10 PM, Cliff Burdick wrote:
>> >
>> > Hi, I'm tr
is
listing?
On Tue, Oct 9, 2018 at 6:54 AM Wiles, Keith wrote:
>
>
> > On Oct 8, 2018, at 11:10 PM, Cliff Burdick wrote:
> >
> > Hi, I'm trying to compile on a machine with an older-generation xeon than
> > the target, so I'm using CONFIG_RTE_MACHINE="broadwell&quo
Hi, I'm trying to compile on a machine with an older-generation xeon than
the target, so I'm using CONFIG_RTE_MACHINE="broadwell" in the config.
gcc's options show that broadwell supports the AES flag, and I verified
that the build shows -march=broadwell. However, when I run my application
it
ion, just remove that action all together
> and check the correctness of your code.
> If all of them are going to q0 (and assuming isolation mode is not
> enabled), then it is most likely because packets don't match any of the
> rte_flow rules you created.
>
> A
>
> On M
Hi, I upgraded to DPDK 18.05 from 17.11 and a previously-working rte_flow
code now fails. I'm using the mlx5 driver, and when I do something similar
to the rte_flow example code, the following error shows:
"VLAN cannot be empty"
I looked at the code in .drivers/net/mlx5/mlx5_flow.c, and it looks
Does anyone know if IBM is working on adding POWER9 support to DPDK? I
noticed there's POWER8 support, but wasn't sure if there were enough
differences to need a new target.
Hi, I have an application where I'm sending a very low rate (1pps) to
certain ports just for gratuitous arp. I allocate a pktmbuf on a different
core, send it through a ring to my tx core thread, and
use rte_eth_tx_burst() to send that packet, and hopefully free. I've
noticed that even though the
Hi, I wanted to do something like when I'm working on a machine that has an
ixgbe device I can load that driver, but when using machine with no special
NIC, just use a generic "PMD" that would go through the Linux stack. Does
such a thing exist? If not, is there a standard way to test dpdk
Hi, I have a DPDK application that I allocate 2048 2MB hugepages on
startup, and everything works for a while. However, if I clean my code and
run the application again, on occasion, I get the following error:
Current CONFIG_RTE_MAX_MEMSEG=256 is not enough
Please either increase it or request
Hi, if I have my own logging mechanism, is there a way to capture the EAL
output, especially on startup? The best way I can think of is to open a
pipe to a child process that reads from stdout, and logs from there, but
that seems like somewhat of a hack. I've seen patches that seem like they
may
Are you including the .mk files? I believe this value is passed in as a
compiled -D flag or defined in rte_memory.h. Do the examples compile
correctly?
On Thu, Jun 23, 2016 at 1:01 AM, SAKTHIVEL ANAND S
wrote:
> hello
>
> I am getting the below error , while compiling my code.. can you pls
Hi, I've been playing around with the sample KNI app, and for some reason
it's extremely slow to respond to pings after the application starts. I'm
starting it with:
./build/kni -c 0xf0 -n 4 -- -p 0x3 -P --config="(0,4,6),(1,5,7)"
and after it starts up, I'm able to ifconfig vEth0 with an IP
Andriy, which pktmbuf_pool do you use when initializing KNI? There's a
separate pool for each queue on a port.
On Mon, Apr 4, 2016 at 9:11 AM, Andriy Berestovskyy
wrote:
> Hi Jason,
>
> On Fri, Mar 25, 2016 at 1:00 AM, Jason Kwon wrote:
> > Is there a KNI example for devices and
> > drivers
Ferruh, sorry.
On Wed, Jun 8, 2016 at 12:48 PM, Cliff Burdick wrote:
> Hi Yigit, when you say "I guess it's possible" is that not common? I would
> think that the amount of traffic people would want to forward to Linux for
> normal DPDK applications would be quite small.
l.
On Wed, Jun 8, 2016 at 12:31 PM, Ferruh Yigit
wrote:
> On 6/8/2016 5:30 PM, Cliff Burdick wrote:
> > Hi, I have an application with two sockets where each core I'm planning
> to
> > transmit and receive a fairly large amount of traffic per core. Each core
> > right now h
you likely don't need a dedicated core
> for the kernel-side RX and TX work.
>
> --
> Matt Laswell
> Principal Software Engineer
> infinite io
>
> On Wed, Jun 8, 2016 at 11:30 AM, Cliff Burdick wrote:
>
>> Hi, I have an application with two sockets where each co
I don't see the answer to this in the documentation, so I'll try here. Why
function does the coremask that's required in rte_eal_init() actually do
other than tell the API what cores you're using? In other words, can't I
just set the core mask to all of the cores in my system, and selectively
use
02:45:10PM -0700, Cliff Burdick wrote:
> > Hi, I've been trying to create a standalone application linked against
> > DPDK, and I'm having some issues getting the program to run properly. The
> > main problem is that in my standalone app compared to the example code,
> > rte_
88 matches
Mail list logo