[dpdk-dev] [RFC 0/4] Use Google Test as DPDK unit test framework

> On Aug 4, 2016, at 2:47 PM, Jim Murphy wrote: > > Hi, > > We are looking at using our existing test environment for our DPDK > applications that will run on our build servers. Hughpages therefore is an > issue. What is involved in running DPDK without huge pages? Command line option

[dpdk-dev] [PATCH] doc: announce ivshmem support removal

2016-08-04 17:00, : > Hi Thomas, > > I'm going to develop a NFV based carrier system with SPP because it's able > to chain VMs in high-performance and I think it might be the best solution > for service function chaining. Without ivshmem, throuput between VMs is > largely decreased. So we are

[dpdk-dev] rte_eth_dev_attach returns 0, although device is not attached

> 4 ???. 2016 ?., ? 16:21, Ferruh Yigit ???(?): > > On 8/4/2016 12:51 PM, Igor Ryzhov wrote: >> Hello Ferruh, >> >>> 4 ???. 2016 ?., ? 14:33, Ferruh Yigit >>> ???(?): >>> >>> Hi Igor, >>> >>> On 8/3/2016 5:58 PM, Igor Ryzhov wrote: Hello. Function rte_eth_dev_attach

[dpdk-dev] [PATCH] ethtool: remove triple license information

On 8/1/2016 1:17 PM, Christian Ehrhardt wrote: > License information is already in LICENSE.GPL. > Remove two extra copies and change referred filename in the files. > > Signed-off-by: Christian Ehrhardt In patch subject, I think it is better to use "kni: " tag instead of "ethtool:", apart from

[dpdk-dev] dpdk 16.07, issues with rte_mempool_create and rte_kni_alloc()

On 8/1/2016 10:19 PM, Gopakumar Choorakkot Edakkunni wrote: > Well, for my purpose I just ended up creating a seperate/smaller pool > earlier during bootup to try to guarantee its from one memseg. > > But I am assuming that this KNI restriction is something thats "currently" > not fixed and is

[dpdk-dev] [PATCH] crypto/qat: optimisation of request copy

On 04/08/16 13:00, Fiona Trahe wrote: > From: Fiona Trahe > > using rte_mov128 instead of structure assignment to copy > template request from session context into request > > Signed-off-by: Fiona Trahe > > --- > drivers/crypto/qat/qat_crypto.c | 2 +- > 1 file changed, 1 insertion(+), 1

[dpdk-dev] [PATCH] doc: announce ivshmem support removal

Hi Thomas, I'm going to develop a NFV based carrier system with SPP because it's able to chain VMs in high-performance and I think it might be the best solution for service function chaining. Without ivshmem, throuput between VMs is largely decreased. So we are not so happy if ivshmem is

[dpdk-dev] [PATCH] kni: memzone info not required to get mbuf address

Originally mempool->mz is used to get address of the mbuf, but now address get directly from mempool, so mempool->mz information is not required. Fixes: d1d914ebbc25 ("mempool: allocate in several memory chunks by default") Signed-off-by: Ferruh Yigit --- lib/librte_kni/rte_kni.c | 13

[dpdk-dev] rte_eth_dev_attach returns 0, although device is not attached

On 8/4/2016 3:54 PM, Igor Ryzhov wrote: > >> 4 ???. 2016 ?., ? 16:21, Ferruh Yigit > > ???(?): >> >> On 8/4/2016 12:51 PM, Igor Ryzhov wrote: >>> Hello Ferruh, >>> 4 ???. 2016 ?., ? 14:33, Ferruh Yigit >>> > ???(?):

[dpdk-dev] [PATCH v1] crypto/qat: make the session struct variable in size

This patch changes the qat firmware session data structure from a fixed size to a variable size which is dependent on the size of the chosen algorithm. This reduces the amount of bytes which are transferred across PCIe and thus helps to increase qat performance when the accelerator is bound by

[dpdk-dev] [PATCH] ivshmem: remove integration in dpdk

On 07/29/2016 03:28 PM, David Marchand wrote: > Following discussions on the mailing list [1] and since nobody stood up to > implement the necessary cleanups, here is the ivshmem integration removal. > > There is not much to say about this patch, a lot of code is being removed. > The default

[dpdk-dev] Mbuf leak issue with IXGBE in vector mod

Hi, 1. When calling rte_eth_dev_stop mbuf pool is depleted. There appears to be a race condition that occurs when RTE_IXGBE_INC_VECTOR is defined: ixgbe_reset_rx_queue(struct ixgbe_adapter *adapter, struct ixgbe_rx_queue *rxq) { ??. #ifdef RTE_IXGBE_INC_VECTOR rxq->rxrearm_start = 0;

[dpdk-dev] [RFC] Generic flow director/filtering/classification API

On Wed, Aug 03, 2016 at 12:11:56PM -0700, John Fastabend wrote: > [...] > > >> The proposal looks very good. It satisfies most of the features > >> supported by Chelsio NICs. We are looking for suggestions on exposing > >> more additional features supported by Chelsio NICs via this

[dpdk-dev] [RFC] Generic flow director/filtering/classification API

On Wed, Aug 03, 2016 at 11:10:49AM -0700, John Fastabend wrote: > [...] > > Considering that allowed pattern/actions combinations cannot be known in > advance and would result in an unpractically large number of > capabilities to > expose, a method is provided to validate a

[dpdk-dev] [PATCH] pci: fix one device probing

The rte_eal_pci_probe_one function could return false positive result if no driver is found for the device. Signed-off-by: Igor Ryzhov --- lib/librte_eal/common/eal_common_pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/librte_eal/common/eal_common_pci.c

[dpdk-dev] rte_eth_dev_attach returns 0, although device is not attached

On 8/4/2016 12:51 PM, Igor Ryzhov wrote: > Hello Ferruh, > >> 4 ???. 2016 ?., ? 14:33, Ferruh Yigit ???(?): >> >> Hi Igor, >> >> On 8/3/2016 5:58 PM, Igor Ryzhov wrote: >>> Hello. >>> >>> Function rte_eth_dev_attach can return false positive result. >>> It happens because

[dpdk-dev] how to design high performance QoS support for a large amount of subscribers

Thank you very much Cristian for the insightful response. Very much appreciated. Regards, Yuyong -Original Message- From: Dumitrescu, Cristian [mailto:cristian.dumitre...@intel.com] Sent: Thursday, August 4, 2016 9:01 AM To: Yuyong Zhang ; dev at dpdk.org; users at dpdk.org Subject:

[dpdk-dev] [PATCH 4/4] doc: make the devbind man page be part of section 8

As a root only program in sbin it should belong to section 8 "8 System administration commands (usually only for root)" Signed-off-by: Christian Ehrhardt --- doc/guides/conf.py | 2 +- mk/rte.sdkinstall.mk | 5 + 2 files changed, 6 insertions(+), 1 deletion(-) diff --git

[dpdk-dev] [PATCH 3/4] doc: add basic invocation info for dpdk-devbind

This summarizes the "how to call dpdk-pmdinfo" in one place to be picked up by html/pdf/man-page docs. That knowledge was available before but spread in various docs along examples (which are great and have to be kept) as well as in the --usage/--help option of the tool itself. Signed-off-by:

[dpdk-dev] [PATCH 2/4] doc: add basic invocation info for dpdk-pmdinfo

This summarizes the "how to call dpdk-pmdinfo" in one place to be picked up by html/pdf/man-page docs. Signed-off-by: Christian Ehrhardt --- doc/guides/conf.py | 4 ++- doc/guides/sample_app_ug/index.rst | 1 + doc/guides/sample_app_ug/pmdinfo.rst | 62

[dpdk-dev] [PATCH 1/4] doc: rendering and installation of man pages

This enables the rendering of rst into man pages as well as installing them (if built) along the binaries. To do so there is a new make target "doc-guides-man" which will render the rst files into man format. Currently these three tools had docs that were compatible "enough" to make up for a

[dpdk-dev] [PATCH 0/4] provide man pages for binaries provided by DPDK

Hi, this is about providing manpages for the binaries installed by DPDK. Eventually people using commands expect at least something reasonable avalable behind "man command". Still it is a try to stick to the rst/sphinx based doc creation. I found that for three of the 5 binaries that are usually

[dpdk-dev] how to design high performance QoS support for a large amount of subscribers

Hi Yuyong, > -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Yuyong Zhang > Sent: Tuesday, August 2, 2016 4:26 PM > To: dev at dpdk.org; users at dpdk.org > Subject: [dpdk-dev] how to design high performance QoS support for a large > amount of subscribers > >

[dpdk-dev] [PATCH] crypto/qat: optimisation of request copy

From: Fiona Trahe using rte_mov128 instead of structure assignment to copy template request from session context into request Signed-off-by: Fiona Trahe --- drivers/crypto/qat/qat_crypto.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

[dpdk-dev] [RFC 0/4] Use Google Test as DPDK unit test framework

Hi, We are looking at using our existing test environment for our DPDK applications that will run on our build servers. Hughpages therefore is an issue. What is involved in running DPDK without hugepages? Thanks, Jim On Wed, Aug 3, 2016 at 1:46 PM, Ming Zhao(??) wrote: > googletest is a

[dpdk-dev] rte_eth_dev_attach returns 0, although device is not attached

Hi Igor, On 8/3/2016 5:58 PM, Igor Ryzhov wrote: > Hello. > > Function rte_eth_dev_attach can return false positive result. > It happens because rte_eal_pci_probe_one returns zero if no driver is found > for the device: > ret = pci_probe_all_drivers(dev); > if (ret < 0) > goto err_return;

[dpdk-dev] limited tx performance with VMware VM + PCI pass-through

Hello Grisha, I'm not sure which OS you are using. We encountered a problem moving from sandybridge based machines to haswell based machines, with using pass-through to VMs. We had to change to a newer version of our OS since the one we used was too old to understand the haswell. We

[dpdk-dev] [PATCH] qat: change the session structure to be variable sized

On 02/08/16 09:30, John Griffin wrote: > This patch changes the qat session data structure sent to qat from a > fixed size to a variable size which is dependent on the size of > the chosen algorithm. > This reduces the amount of bytes which are transferred across > PCIe and thus helps to increase

[dpdk-dev] Using qemu-system-x86_64 of KVM sourceforge project versus qemu-system-x86_64 of http://www.qemu.org/

Hi Mauricio, Thanks! >be sure that kvm is enabled by > setting "accel=kvm" in the qemu command line. Isn't "--enable-kvm" intended for that ? Kevin

[dpdk-dev] [PATCH 2/2] net/bonding: enable slave VLAN filter

SR-IOV virtual functions cannot rely on promiscuous mode for the reception of VLAN tagged frames. Program the vlan filter for each slave when a vlan is configured for the bonding master. Signed-off-by: Eric Kinzie --- drivers/net/bonding/rte_eth_bond_api.c | 68

[dpdk-dev] [PATCH 1/2] net/bonding: validate speed after link up

It's possible for the bonding driver to mistakenly reject an interface based in it's, as yet, unnegotiated link speed and duplex. Always allow the interface to be added to the bonding interface but require link properties validation to succeed before slave is activated. Fixes: 2efb58cbab6e

[dpdk-dev] [PATCH 0/2] bonding link validation and vlan filters

This series contains a fix to the validation of interfaces added to a bond group and an enhancement to allow reception of tagged frames on slave interfaces that are virtual functions. Eric Kinzie (2): net/bonding: validate speed after link up net/bonding: enable slave VLAN filter

[dpdk-dev] Using qemu-system-x86_64 of KVM sourceforge project versus qemu-system-x86_64 of http://www.qemu.org/

On 08/04/2016 10:27 AM, Kevin Wilson wrote: > Hi Mauricio, > Thanks! > >> be sure that kvm is enabled by >> setting "accel=kvm" in the qemu command line. > Isn't "--enable-kvm" intended for that ? Sorry, I meant "-machine accel=kvm". I don't know what is the exact difference but enable-kvm

[dpdk-dev] Running DPDK in a VM

Hi! I am looking at running DPDK in a VM, I would like to know if there are any limitations when doing this in terms of any DPDK features that do not work in a VM The only post I came across for running DPDK in a VM is this: http://dpdk.org/ml/archives/dev/2013-September/000441.html As an

[dpdk-dev] Using qemu-system-x86_64 of KVM sourceforge project versus qemu-system-x86_64 of http://www.qemu.org/

Hi Kevin, On 08/04/2016 08:55 AM, Kevin Wilson wrote: > Hi, > I am trying to use DPDK SRIOV passthrough with DPDK on Intel NICs. > I am following the instructions in "Network Interface Controller Drivers", in > http://fast.dpdk.org/doc/pdf-guides/nics-16.07.pdf > > I saw in "11.2 Setting Up a

[dpdk-dev] FW: DPDK Community Survey - about to close

Hi all Final reminder that the survey closes today Thanks to those who have already provided their input! Regards Mike -Original Message- From: Thomas Monjalon [mailto:thomas.monja...@6wind.com] Sent: Tuesday, August 2, 2016 1:45 PM To: users at dpdk.org; dev at dpdk.org Cc: Glynn,

[dpdk-dev] [PATCH] examples/exception_path: fix shift operation in lcore setup

On 8/3/2016 12:44 PM, Daniel Mrzyglod wrote: > The operaton may have an undefined behavior or yield to an unexpected result. > A bit shift operation has a shift amount which is too large or has a negative > value. > > Coverity issue: 30688 > Fixes: ea977ff1cb0b ("examples/exception_path: fix

[dpdk-dev] Using qemu-system-x86_64 of KVM sourceforge project versus qemu-system-x86_64 of http://www.qemu.org/

Hi, I am trying to use DPDK SRIOV passthrough with DPDK on Intel NICs. I am following the instructions in "Network Interface Controller Drivers", in http://fast.dpdk.org/doc/pdf-guides/nics-16.07.pdf I saw in "11.2 Setting Up a KVM Virtual Machine Monitor" in this pdf that it says to download

[dpdk-dev] [PATCH] app/testpmd: fix RSS-hash-key size

Hi Awal, As I said in the offline discussion, here are few nits that I think need to be addressed. See below. Konstantin > > RSS hash-key-size is retrieved from device configuration instead of using a > fixed size of 40 bytes. > > Fixes: f79959ea1504 ("app/testpmd: allow to configure RSS

[dpdk-dev] [PATCH 2/2] examples/tep_term: fix inner L4 checksum failure

When sending packets from virtual machine which in need of TSO by hardware NIC, the inner L4 checksum is not correct on the other side of the cable. It's because get_psd_sum() depends on PKT_TX_TCP_SEG to calculate pseudo-header checksum, but currently this bit is set after the function

[dpdk-dev] [PATCH 1/2] examples/tep_term: fix offload on VXLAN failure

Based on previous fix of offload on VXLAN using i40e, applications need to set proper tunneling type on ol_flags so that i40e driver can pass it to NIC. Fixes: a50245ede72a ("examples/tep_term: initialize VXLAN sample") Signed-off-by: Jianfeng Tan --- examples/tep_termination/vxlan.c | 2 ++ 1

[dpdk-dev] [PATCH 0/2] Two offloading issues of tep_term

This patch set depends on: - http://dpdk.org/ml/archives/dev/2016-August/044924.html Patch 1: fill tunneling type. Patch 2: inner L4 checksum error. Signed-off-by: Jianfeng Tan Jianfeng Tan (2): examples/tep_term: fix offload on VXLAN failure examples/tep_term: fix inner L4 checksum