[dpdk-dev] Build failure on Ubuntu 14.04

Hi Keith, we built the newest dpdk code on my machine, it seems OK, please see UB14.04 info in my computer. System: Ubuntu14.04 Kernel: 3.13.0-24 X86_64 Compiler: GCC 4.8.2 x86_64 Can you let me know which kernel version you use? Thanks Waterman Original Message- >From: d

[dpdk-dev] Build failure on Ubuntu 14.04

> > Hi Keith, > > we built the newest dpdk code on my machine, it seems OK, > please see UB14.04 info in my computer. > System: Ubuntu14.04 > Kernel: 3.13.0-24 X86_64 > Compiler: GCC 4.8.2 x86_64 > > Can you let me know which kernel version you use? Apparently, it is this one: 3

[dpdk-dev] Testing memnic for VM to VM transfer

Hi Thomas: Do we have recommendation between virtio-PMD and memnic-PMD? On Thu, Jul 3, 2014 at 8:05 PM, Thomas Monjalon wrote: > 2014-07-03 19:03, GongJinrong: >> Hi, Thomas, I cannot find the host-sim, where can I get it? > > In memnic repository: > http://dpdk.org/browse/memnic/commit

[dpdk-dev] Testing memnic for VM to VM transfer

Hi, 2014-07-07 17:58, xie huawei: > Hi Thomas: > > Do we have recommendation between virtio-PMD and memnic-PMD? I don't understand your question. I think your question is not related to the test application for memnic. Please, could you open a new thread for this question (more detailed)? -- T

[dpdk-dev] Making space in mbuf data-structure

Hello Bruce, Thank you to revive this discussion now that the 1.7 is released. First, I would like to reference my previous patch series that first reworks the mbuf to gain 9 bytes [1]. The v1 of the patch was discussed at [2]. Now, let's list what I would find useful to have in this mbuf rework

[dpdk-dev] Build failure on Ubuntu 14.04

Someone compiling on CentOS kernel 3.13.7 needed the original code. Looks like some other type of ifdef change needs to be done to the kcompat.h file to allow it to compile on Ubuntu 14.04 with kernel 3.13.0-30. Keith Wiles, Principal Technologist with CTO office, Wind River mobile 972-213-5533

[dpdk-dev] Build failure on Ubuntu 14.04

The guy using CentOS stated if he changed the code to compare with KERNEL_VERSION(3,13,8) it works. Also the compiler is older on CentOS gcc 4.4.7 and has a problem with the: "--param max-inline-insns-single=100" issue in librte_eal/linuxapp/igb_uio/Makefile Giving the error: "error: inlining f

[dpdk-dev] Build failure on Ubuntu 14.04

BTW, he changed the value from 100 to 400 and that fixed his compile problem on CentOS. Keith Wiles, Principal Technologist with CTO office, Wind River mobile 972-213-5533 [Powering 30 Years of Innovation] On Jul 7, 2014, at 9:59 AM, Wiles, Roger Keith

[dpdk-dev] Performance issue with vmxnet3 pmd

Hi, I was testing l2-fwd with vmxnet3 pmd (included in dpdk). The maximum forwarding rate I got from vmxnet3 pmd with l2fwd is only 2.5 to 2.8 Gbps. This is in contrast with ixgbe pmd with which I could easily achieve 10 gbps forwarding rate. With the original vmxnet3 driver (non pmd), I could

[dpdk-dev] Performance issue with vmxnet3 pmd

Hi Hyunseok, We should not compare Vmxnet3-PMD with ixgbe-PMD performance as Vmxnet3 device is a para-virtual device and it's not similar to directly assigned device to a VM either. There is VMEXIT/VMEXIT occurrence at burst-size boundary and that overhead can?t be eliminated unless the design

[dpdk-dev] Performance issue with vmxnet3 pmd

Thanks for your response. I am actually more interested in stock (non-dpdk) vmxnet3 driver vs. vmxnet3 pmd driver comparison. When I forward pkts from stock vmxnet3 driver, I am able to achieve much higher throughput than with vmxnet3 pmd. To make comparison fair, I did not leverage gro/gso. Do