>Hello, I've encountered a problem with running ovs with IVSHMEM.
>
>I've followed the guide of INSTALL.DPDK.md of ovs-2.5.0 to use IVSHMEM.
>
>Firstly, I started ovs-vswitchd using "./sbin/ovs-vswitchd --dpdk -c 0x1 -n 4
>--
>proc-type=primary -- --pidfile --detach", and I added dpdk
>ports(dpdk0
>Hi:
>When I read the code of emc_cache in dips-netdev.c, I could not understand
>the replacement policy below in emc_insert() as follows():
> /* Replacement policy: put the flow in an empty (not alive) entry, or
> * in the first entry where it can be */
> if (!to_be_replaced
>
>
>Another question is:
>Why “-numa” option is needed to start the VM?
It's worth referring this link http://download.qemu.org/qemu-doc.html for all
the information on the qemu arguments.
- Bhanuprakash.
>
>Wang Zhike
>
>From: 王志克
>Sent: Monday, July 03, 2017 9:06 PM
>To: ovs-discuss@openvsw
>>What is your use case(s) ?
>>My usecase might be setup a VBRAS VNF with OVS-DPDK as an NFV normal
>>case, and it requires a good performance, however, OVS-DPDK seems still
>not
>>reach its needs compared with hardware offloading, we are evaluating VPP
>as
>>well,
>As you mentioned VPP here, It's
It’s a long weekend in US and will try answering some of your questions in
Darrell's absence.
>Why do think having more than 64k per PMD would be optimal ?
>I originally thought that the bottleneck in classifier because it is saturated
>full
>so that look up has to be going to flow table, so I t
tls.
>
>You are probably enabled hyper-thread with 35MB and got 28 cores.
I have E5-2695 v3, dual socket with 14 cores per socket. I will have 56 cores
with HT enabled.
- Bhanuprakash.
>
>[1] https://mail.openvswitch.org/pipermail/ovs-dev/2015-May/298999.html
>
>
>
>O
>
>I guess the answer is now the general LLC is 2.5M per core so that there is 64k
>flows per thread.
AFAIK, the no. of flows here may not have to do anything with LLC. Also there
is EMC cache(8k entries) of ~4MB per PMD thread.
Yes the performance will be nice with simple test cases (P2P with 1
>Question: what are the additional overhead in the standard OVS that cause
>to poor performance related to the OVS-DPDK setup ?
>I'm not talking about the PMD improvements (OVS-DPDK) running on the
>host - but on overhead in the VM context in the standard OVS setup
When running guest instance
>Hi All,
>My requirement is my application should detect the crash in ovsdb-server and
>ovs-vswitchd application and do required relaunching applications.
Have you checked '--monitor' option? With that option the ovs-vswitchd will be
relaunched automatically in case of signal.
Please add subject
>Hi,
>
>I've a question regarding how OVS utilizes the CPU. Does OVS supports multi-
>core? My understanding is kernel datapath module is multi-core but
>userspace module is single core. is this correct?
Not really, I would assume you are asking about OvS-DPDK that uses DPDK
datapath. It implem
Did you Delete the datapath? Try removing the module after deleting the
datapath.
$ ovs-dpctl del-dp system@ovs-system
Bhanuprakash.
From: ovs-discuss-boun...@openvswitch.org
[mailto:ovs-discuss-boun...@openvswitch.org] On Behalf Of ???
Sent: Wednesday, March 29, 2017 9:26 AM
To: ovs-discuss
S
han IGB_UIO +
>more forgiving [1]. Hence switched to the same.
>
>[1] http://dpdk-guide.gitlab.io/dpdk-guide/setup/binding.html
>
>/Shivaram
>
>On Tue, Mar 28, 2017 at 1:12 PM, Bodireddy, Bhanuprakash
> wrote:
>
>>any hints for me based on the info provided?
>
>Your B
device
>:00:1a.0 [0x7ba75000 - 0x7ba84fff]
>[ 1.689028] DMAR: Ignoring identity map for HW passthrough device
>:00:1d.0 [0x7ba75000 - 0x7ba84fff]
>[ 1.689036] DMAR: Prepare 0-16MiB unity mapping for LPC
>[ 1.689036] DMAR: Ignoring identity map for HW passthrough device
&
Hello shivaram,
Darrell has already asked if VT-d is enabled?
Your cmdline options are appropriate as you have iommu enabled(intel_iommu=on
iommu=pt).
can you also check the output of below mentioned command to see if VT-d is
enabled in BIOS?
$ dmesg | grep -e DMAR -e IOMMU
Regards,
Bhanupraka
>-Original Message-
>From: Advith Nagappa [mailto:advith.naga...@gmail.com]
>Sent: Monday, March 27, 2017 9:45 AM
>To: Bodireddy, Bhanuprakash
>Subject: Re: [ovs-discuss] Userspace OVS - Kernel Hashtable
>
>Hi,
>
>Thank you for the response!
>
>Is ther
>-Original Message-
>From: ovs-discuss-boun...@openvswitch.org [mailto:ovs-discuss-
>boun...@openvswitch.org] On Behalf Of Advith Nagappa
>Sent: Monday, March 27, 2017 7:03 AM
>To: disc...@openvswitch.org
>Subject: [ovs-discuss] Userspace OVS - Kernel Hashtable
>
>Hello All,
>
>Does a purel
>Hi all
>
>I need to specify a number of classes of service according to DSCP bits. Each
>class of service should be associated with a priority. For example: Packet A is
>sent with a DSCP value of 0 and packet B is sent with a value of 46; 0 has a
>lower priority than 46. Packet A arrives before pa
>-Original Message-
>From: ovs-discuss-boun...@openvswitch.org [mailto:ovs-discuss-
>boun...@openvswitch.org] On Behalf Of Patrick Adler
>Sent: Tuesday, March 14, 2017 8:50 AM
>To: ovs-discuss@openvswitch.org
>Subject: [ovs-discuss] Configuring openvswitch dpdk with "--file-prefix"
>
>Hi,
>
>-Original Message-
>From: ovs-discuss-boun...@openvswitch.org [mailto:ovs-discuss-
>boun...@openvswitch.org] On Behalf Of Priyanka
>Sent: Friday, March 3, 2017 9:17 AM
>To: ovs-discuss@openvswitch.org
>Subject: [ovs-discuss] Inter VM communication using ovs-dpdk
>
>Hi,
>
>We have a OVS-DPD
>-Original Message-
>From: ovs-discuss-boun...@openvswitch.org [mailto:ovs-discuss-
>boun...@openvswitch.org] On Behalf Of Kumuthini Ratnasingham
>Sent: Thursday, February 9, 2017 12:14 AM
>To: disc...@openvswitch.org
>Subject: [ovs-discuss] OVS Packet Per Second Performance
>
>Hi there,
>
>Subject: [ovs-discuss] RPM from DPDK and OVS
>
>Hi!
>
>I want to build a custom DPDK and a custom OVS into rpm. I'm on CentOs7
>latest. Can you refer a good guide for me?
This should be helpful, also there is a link internally to RHEL7/derivatives
like CentOS7.
http://docs.openvswitch.org/en/lat
This might be a good idea in some scenarios and I have come across this
earlier. We have done this in eNB to immunize the key daemon from oom killer by
setting oom_adj of the process to OOM_DISABLE.
Regards,
Bhanuprakash.
From: ovs-discuss-boun...@openvswitch.org
[mailto:ovs-discuss-boun...@op
22 matches
Mail list logo
