[dpdk-users] all rx packets are zero on dpdk but function well on linux

2018-05-27 Thread edgar helmut
Hi, I have a 82599ES that functions well on linux (i.e. ixgbe). However, when binded to dpdk 100% of the received packets are counted as missed (using testpmd as well as my own stable application), and non arrive to the application. I wonder what can be the reason. Why does it functions well on l

Re: [dpdk-users] Dpdk poor performance on virtual machine

2017-01-03 Thread edgar helmut
e to count pkts in/out to and from the vm. If someone has an idea how to do that i will be more than happy to hear. edgar On Wed, Dec 28, 2016 at 10:09 AM edgar helmut wrote: > I tried this procedure as well as few others. > at all of the procedures the HugePages_Free or HugePages

Re: [dpdk-users] Dpdk poor performance on virtual machine

2016-12-28 Thread edgar helmut
esn't really make the job. any idea? On Tue, Dec 27, 2016 at 8:52 PM Stephen Hemminger < step...@networkplumber.org> wrote: > On Tue, 27 Dec 2016 15:59:08 + > edgar helmut wrote: > > > short explanation for how to read the comparison: > > first row is packet

Re: [dpdk-users] Dpdk poor performance on virtual machine

2016-12-27 Thread edgar helmut
throughput) in gbps. i.e. on 1500 bytes packet size testpmd delivers ~9.82 gbps from port 1 to 2 and another ~9.82 gbps from port 2 to 1, while at the vm it only delivers ~3.9 gbps for each direction. On Tue, Dec 27, 2016 at 5:52 PM edgar helmut wrote: > Thanks. That's the document i am f

Re: [dpdk-users] Dpdk poor performance on virtual machine

2016-12-27 Thread edgar helmut
for to measure in and out packets through host, maybe > you can look at vhost/virtio interface also. > > > > After your testing, if you can report the performace out with macvtap, > that also helps us. J > > > > > > *From:* edgar helmut [mailto:helmut.edgar...@gmail.

Re: [dpdk-users] Dpdk poor performance on virtual machine

2016-12-24 Thread edgar helmut
t; huge pages and macvtap. Any specific purpose? > > > > I think you should get a baseline first, then to get how much perf dropped > if using anonymous hugepages or macvtap。 > > 1. Baseline: real hugepage + real pci pass-through > > 2. Anon hugepages vs hugepages

Re: [dpdk-users] Dpdk poor performance on virtual machine

2016-12-24 Thread edgar helmut
perf dropped if using anonymous hugepages or macvtap。 1. Baseline: real hugepage + real pci pass-through 2. Anon hugepages vs hugepages 3. Real pci pass-through vs. macvtap *From:* edgar helmut [mailto:helmut.edgar...@gmail.com] *Sent:* Saturday, December 24, 2016 3:23 AM *To:*

Re: [dpdk-users] Dpdk poor performance on virtual machine

2016-12-23 Thread edgar helmut
that forwarding ~5-6 gbps between two ports should be trivial, so it will be great if someone can share its configuration for a tested setup. Any further idea will be highly appreciated. Thanks. On Sat, Dec 17, 2016 at 2:56 PM edgar helmut wrote: > That's what I afraid. > In fact i need

Re: [dpdk-users] Dpdk poor performance on virtual machine

2016-12-17 Thread edgar helmut
d ~4G > (4360192KB). So some were mapped to 4K pages. > > > > BTW, the memory used by transparent hugepage is not the hugepage you > reserved in kernel boot option. > > > > *From:* edgar helmut [mailto:helmut.edgar...@gmail.com] > *Sent:* Friday, December 16, 201

Re: [dpdk-users] Dpdk poor performance on virtual machine

2016-12-15 Thread edgar helmut
x27;t grantee the app is > using > > the real huge pages. > > > > > > -Original Message----- > > From: users [mailto:users-boun...@dpdk.org] On Behalf Of edgar helmut > > Sent: Thursday, December 15, 2016 9:32 PM > > To: Wiles, Keith > > Cc: use

Re: [dpdk-users] Dpdk poor performance on virtual machine

2016-12-15 Thread edgar helmut
> Sometimes, transparent huge page mechanism doesn't grantee the app is using > the real huge pages. > > > -Original Message- > From: users [mailto:users-boun...@dpdk.org] On Behalf Of edgar helmut > Sent: Thursday, December 15, 2016 9:32 PM > To: Wiles, K

Re: [dpdk-users] Dpdk poor performance on virtual machine

2016-12-15 Thread edgar helmut
see significant difference between the vm and the host which I can't fully explain. any further idea? Regards, Edgar On Thu, Dec 15, 2016 at 2:54 PM, Wiles, Keith wrote: > > > On Dec 15, 2016, at 1:20 AM, edgar helmut > wrote: > > > > Hi. > > Some help is nee

[dpdk-users] Dpdk poor performance on virtual machine

2016-12-14 Thread edgar helmut
Hi. Some help is needed to understand performance issue on virtual machine. Running testpmd over the host functions well (testpmd forwards 10g between two 82599 ports). However same application running on a virtual machine over same host results with huge degradation in performance. The testpmd th

Re: [dpdk-users] new_device is never created at the vhost-switch sample app

2016-12-14 Thread edgar helmut
install...). Thanks On 05 Dec 2016 7:27 PM, "gowrishankar muthukrishnan" < gowrishanka...@linux.vnet.ibm.com> wrote: On Friday 25 November 2016 01:28 PM, edgar helmut wrote: > Hi, > I am following http://dpdk.org/doc/guides/sample_app_ug/vhost.html using > 16.11 in

Re: [dpdk-users] new_device is never created at the vhost-switch sample app

2016-12-03 Thread edgar helmut
Hi, the silence confuses me, at least I would like to understand if it is well supported or not for the release so I can decide whether to find an alternative or not. what is the trigger for the guest to send the VHOST_USER_SET_VRING_KICK? Thanks Edgar On Fri, Nov 25, 2016 at 9:58 AM edgar

[dpdk-users] new_device is never created at the vhost-switch sample app

2016-11-25 Thread edgar helmut
Hi, I am following http://dpdk.org/doc/guides/sample_app_ug/vhost.html using 16.11 in order to have a simple VM forwarding packets from one NIC to another efficiently, but it doesn't work... so I need some help to understand what am i missing. I am using x520 (intel 82599). host is ubuntu 16.04 and