[dpdk-dev] PEG6I Six Port Gigabit

Hi, has anyone tested dpdk on this card? PEG6I Six Port Gigabit http://www.servethehome.com/dell-silicom-peg6i-port-gigabit-gige-intel-82571eb-network-adapter/ It is an 82571EB chip. Regards

[dpdk-dev] plumbing say test-pmd for non-Intel NICs(drivers)

Hello, 29/05/2013 10:27, somnath kotur : > Pls pardon me if this has been asked/addressed in another thread. Just > wanted to know what i need to do if i wanted to plumb my non-Intel 10G NIC > into the DPDK , For instance just into the test-pmd framework would also be > fine ? > Any pointers /ti

[dpdk-dev] plumbing say test-pmd for non-Intel NICs(drivers)

On 29.05.2013 10:27, somnath kotur wrote: > Hi, >Pls pardon me if this has been asked/addressed in another thread. Just > wanted to know what i need to do if i wanted to plumb my non-Intel 10G NIC > into the DPDK , For instance just into the test-pmd framework would also be > fine ? > Any point

[dpdk-dev] Best example for showing throughput?

On 05/28/2013 09:15 PM, Patrick Mahan wrote: > So the overhead cost is almost 70%? > > Can this ever do line rate? Under what conditions? It has been my experience > that the industry standard is testing throughput using these 64 byte packets. This overhead can actually be explained considering

[dpdk-dev] plumbing say test-pmd for non-Intel NICs(drivers)

part -- An HTML attachment was scrubbed... URL: <http://dpdk.org/ml/archives/dev/attachments/20130529/c2249b22/attachment.html>

[dpdk-dev] Performances are not scale with multiple ports

Hi shiane and all Emre and Stephen, thanks your suggest for us. We are not testing multi-port traffic gen using NETMAP though DPDK l3fw, but we understood some packet recive-side without NETMAP Linux. Please try it using stephen's pktgen if you have more traffic gen/recive server resources or Us

[dpdk-dev] Best example for showing throughput?

On May 29, 2013, at 7:07 AM, Damien Millescamps wrote: > On 05/28/2013 09:15 PM, Patrick Mahan wrote: >> So the overhead cost is almost 70%? >> >> Can this ever do line rate? Under what conditions? It has been my >> experience that the industry standard is testing throughput using these 64 >

[dpdk-dev] plumbing say test-pmd for non-Intel NICs(drivers)

On Wed, 29 May 2013 16:23:07 +0200 Antti Kantee wrote: > On 29.05.2013 10:27, somnath kotur wrote: > > Hi, > >Pls pardon me if this has been asked/addressed in another thread. Just > > wanted to know what i need to do if i wanted to plumb my non-Intel 10G NIC > > into the DPDK , For instance

[dpdk-dev] [PATCH v2 3/3] mk: rule help

The goal of this rule is to show which rules/options are available from a "make" command. Let's start by printing short doc about SDK building. Signed-off-by: Thomas Monjalon --- mk/rte.sdkdoc.mk |5 + mk/rte.sdkroot.mk |3 ++- 2 files changed, 7 insertions(+), 1 deletion(-) diff -

[dpdk-dev] [PATCH v2 2/3] mk: rule showconfigs for configs listing

Allow to list configs in config/ directory with a simple make rule. The rule showconfigs is reused in config error. In the same time, "echo -n" is replaced by the more portable "printf". Signed-off-by: Thomas Monjalon --- mk/rte.sdkconfig.mk | 12 +++- mk/rte.sdkroot.mk |4 ++--

[dpdk-dev] [PATCH v2 1/3] doc: how to build

This is a cheat sheet to build DPDK and can be used for a "make help". It is explicitly described as a build help in order to concatenate it with other helps such as test commands. Signed-off-by: Thomas Monjalon --- doc/build-sdk-quick.txt | 20 1 file changed, 20 insertio

[dpdk-dev] [PATCH v2 0/3] make help

Let's add help for building DPDK. changes since v1: (Olivier's review) - reword/reorganize help - reuse "showconfigs" in "config" - no magic sed to call "make showconfigs" in help --- Thomas Monjalon (3): doc: how to build mk: rule showconfigs for configs listing mk: rule help doc/build-