Re: [dpdk-dev] [PATCH V3 5/5] app/testpmd: cli for traffic metering and policing

> -Original Message- > From: Dumitrescu, Cristian > Sent: Friday, October 6, 2017 10:46 PM > To: dev@dpdk.org > Cc: tho...@monjalon.net; adrien.mazarg...@6wind.com; Wu, Jingjing > ; hemant.agra...@nxp.com; > jerin.ja...@caviumnetworks.com; Singh, Jasvinder > Subject: [PATCH V3 5/5] app/t

Re: [dpdk-dev] [PATCH v7 1/3] app/test-pmd: add CLI for TM capability and stats

Tested testpmd CLI for TM with ixgbe, i40e and softnic. Tested-by: Yulong Pei -Original Message- From: Singh, Jasvinder Sent: Wednesday, October 11, 2017 5:26 PM To: dev@dpdk.org Cc: Dumitrescu, Cristian ; Wu, Jingjing ; De Lara Guarch, Pablo ; Pei, Yulong Subject: [PATCH v7 1/3] a

Re: [dpdk-dev] [PATCH v2 3/4] event/dpaa2: support event eth adapter

On 10/12/2017 10:48 PM, Nipun Gupta wrote: Signed-off-by: Nipun Gupta --- drivers/event/Makefile | 4 +- drivers/event/dpaa2/Makefile | 2 + drivers/event/dpaa2/dpaa2_eventdev.c | 151 ++- drivers/event/dpaa2/dpaa2_eventdev.h | 8 +

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

Hi Jerin, The patch is as below. Jie and I just modified from the original case. 1. If all consumers(slave lcore) start to work after the producer fill the ring in every iteration test, there will be no problem. 2. If only one consumer with one producer, there will not be any problem eithe

Re: [dpdk-dev] Vendor specific sub-trees under next-net

Hi Ferruh, On 10/13/2017 5:21 AM, Thomas Monjalon wrote: 13/10/2017 01:31, Ferruh Yigit: Hi Thomas, et al Previously it has been mentioned [1] to have vendor specific driver trees under next-net. And recently Mellanox agreed to have a Mellanox tree [2]. Intel also agrees to have next-net-int

Re: [dpdk-dev] [PATCH 4/4] eal/x86: implement arch-specific TSC freq query

On Fri, 13 Oct 2017 02:02:47 +0200 Thomas Monjalon wrote: > +static uint32_t > +check_model_wsm_nhm(uint8_t model) > +{ > + switch (model) { > + /* Westmere */ > + case 0x25: > + case 0x2C: > + case 0x2F: > + /* Nehalem */ > + case 0x1E: > + case 0x1F: > + case

Re: [dpdk-dev] [PATCH v5 3/4] app/testpmd: add raw flow type to flow director

> break; > + case RTE_ETH_FLOW_RAW: > + raw_packet_buff = open_ddp_package_file(res->filepath, > + &raw_packet_size); If you think open_ddp_package_file can be used for common binary file open, could you rename it an

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

Hi Jerin On 10/13/2017 9:49 AM, Jerin Jacob Wrote: -Original Message- Date: Fri, 13 Oct 2017 09:16:31 +0800 From: Jia He To: Jerin Jacob , "Ananyev, Konstantin" Cc: Olivier MATZ , "dev@dpdk.org" , "jia...@hxt-semitech.com" , "jie2@hxt-semitech.com" , "bing.z...@hxt-semite

Re: [dpdk-dev] [PATCH] igb_uio: fix unknown symbols

On 10/12/2017 10:46 PM, Ferruh Yigit wrote: This patch partially reverts the commit d196343a258e and adds some functions from Markus' previous version of the patch [1]. igb_uio uses pci_msi_unmask_irq() and pci_msi_mask_irq() kernel APIs when kernel version is >= 3.19 because these APIs are impl

Re: [dpdk-dev] [PATCH 0/4] improve tsc frequency calibration

-Original Message- > Date: Fri, 13 Oct 2017 02:02:43 +0200 > From: Thomas Monjalon > To: gowrishanka...@linux.vnet.ibm.com, jerin.ja...@caviumnetworks.com, > jianbo@linaro.org, sergio.gonzalez.mon...@intel.com, > bruce.richard...@intel.com > Cc: dev@dpdk.org > Subject: [PATCH 0/4] im

Re: [dpdk-dev] [PATCH v7 3/3] app/test-pmd: add CLI for TM nodes and hierarchy commit

> + /* Port status */ > + if (port_is_started(port_id)) { > + printf(" Port %u not stopped (error)\n", port_id); > + return; > + } > + > + /* Node parameters */ > + if (res->parent_node_id < 0) > + parent_node_id = UINT32_MAX; > + else > +

[dpdk-dev] [PATCH 5/6] net/i40e: fix not supporting NULL TM profile

It's by design that APP can add a TM node without shaper profile. But i40e doesn't support it currently. Fixes: 03a249b62bbd ("net/i40e: support adding TM node") Fixes: cac29c3c00a4 ("net/i40e: support committing TM hierarchy") Signed-off-by: Wenzhuo Lu --- drivers/net/i40e/i40e_tm.c | 47 ++

[dpdk-dev] [PATCH 6/6] net/ixgbe: fix not supporting NULL TM profile

It's by design that APP can add a TM node without shaper profile. But ixgbe doesn't support it currently. Fixes: e0ff4d304ccf ("net/ixgbe: support adding TM node") Fixes: 5713ade69776 ("net/ixgbe: support committing TM hierarchy") Signed-off-by: Wenzhuo Lu --- drivers/net/ixgbe/ixgbe_tm.c | 40 +

[dpdk-dev] [PATCH 4/6] net/ixgbe: fix TM level capability getting

Only queue nodes should be taken as leaf nodes, all the other nodes are non-leaf nodes. Correct it when getting the TM level capability. Fixes: 596988e193f7 ("net/ixgbe: support getting TM level capability") Signed-off-by: Wenzhuo Lu --- drivers/net/ixgbe/ixgbe_tm.c | 41 +++-

[dpdk-dev] [PATCH 3/6] net/ixgbe: fix TM node parameter checking

Only queue nodes should be taken as leaf nodes, all the other nodes are non-leaf nodes. Correct it when checking the parameters of the TM nodes. Fixes: e0ff4d304ccf ("net/ixgbe: support adding TM node") Signed-off-by: Wenzhuo Lu --- drivers/net/ixgbe/ixgbe_tm.c | 10 +- 1 file changed, 5

[dpdk-dev] [PATCH 1/6] net/i40e: fix TM node parameter checking

Only queue nodes should be taken as leaf nodes, all the other nodes are non-leaf nodes. Correct it when checking the parameters of the TM nodes. Fixes: 03a249b62bbd ("net/i40e: support adding TM node") Signed-off-by: Wenzhuo Lu --- drivers/net/i40e/i40e_tm.c | 12 +++- 1 file changed, 7

[dpdk-dev] [PATCH 2/6] net/i40e: fix TM level capability getting

Only queue nodes should be taken as leaf nodes, all the other nodes are non-leaf nodes. Correct it when getting the TM level capability. Fixes: 0fb1ef1e7930 ("net/i40e: support getting TM level capability") Signed-off-by: Wenzhuo Lu --- drivers/net/i40e/i40e_tm.c | 41 +++

[dpdk-dev] [PATCH 0/6] fix TM (Traffic Management) issues on i40e and ixgbe

By design, all the queues are leaf nodes and others are non-leaf nodes, and the shaper profile can be null. Currently the i40e and ixgbe implementation doesn't follow this design. This patch set fixes this issue. Wenzhuo Lu (6): net/i40e: fix TM node parameter checking net/i40e: fix TM level c

Re: [dpdk-dev] [PATCH] igb_uio: fix unknown symbols

On 10/13/2017 3:46 AM, Ferruh Yigit wrote: > This patch partially reverts the commit d196343a258e and adds some > functions from Markus' previous version of the patch [1]. > > igb_uio uses pci_msi_unmask_irq() and pci_msi_mask_irq() kernel APIs > when kernel version is >= 3.19 because these APIs a

Re: [dpdk-dev] [PATCH v7 5/6] igb_uio: use kernel functions for masking MSI-X

On 10/12/2017 6:04 PM, Ferruh Yigit wrote: > On 10/9/2017 10:56 PM, Patrick MacArthur wrote: >> Hi, Markus, >> >> This commit appears to cause a regression on CentOS 7.4 with the in-box >> Linux 3.10.0-639-2.2.el7.x86_64 kernel. Although the kernel module >> appears to build correctly, when I att

[dpdk-dev] [PATCH] igb_uio: fix unknown symbols

This patch partially reverts the commit d196343a258e and adds some functions from Markus' previous version of the patch [1]. igb_uio uses pci_msi_unmask_irq() and pci_msi_mask_irq() kernel APIs when kernel version is >= 3.19 because these APIs are implemented in this Linux kernel version. But the

Re: [dpdk-dev] [PATCH v7 1/3] app/test-pmd: add CLI for TM capability and stats

> -Original Message- > From: Singh, Jasvinder > Sent: Wednesday, October 11, 2017 5:26 PM > To: dev@dpdk.org > Cc: Dumitrescu, Cristian ; Wu, Jingjing > ; De Lara Guarch, Pablo > ; Pei, Yulong > Subject: [PATCH v7 1/3] app/test-pmd: add CLI for TM capability and stats > > Add following

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

-Original Message- > Date: Fri, 13 Oct 2017 02:12:58 + > From: "Zhao, Bing" > To: "Liu, Jie2" , "Ananyev, Konstantin" > , Olivier MATZ , > "dev@dpdk.org" , "jerin.ja...@caviumnetworks.com" > > CC: "He, Jia" , Jia He > Subject: RE: [PATCH] ring: guarantee ordering of cons/prod load

Re: [dpdk-dev] [PATCH v7 2/3] app/test-pmd: add CLI for shaper and wred profiles

> +/* *** Add Port TM Node WRED Profile *** */ struct > +cmd_add_port_tm_node_wred_profile_result { > + cmdline_fixed_string_t add; > + cmdline_fixed_string_t port; > + cmdline_fixed_string_t tm; > + cmdline_fixed_string_t node; > + cmdline_fixed_string_t wred; > + cmdline_f

Re: [dpdk-dev] [PATCH 0/6] fix port id type

Hi Ferruh, OK, I see. I will send another patches to fix them as soon as possible. Thanks Zhiyong > -Original Message- > From: Yigit, Ferruh > Sent: Friday, October 13, 2017 10:20 AM > To: Yang, Zhiyong ; dev@dpdk.org; > tho...@monjalon.net; Richardson, Bruce > Subject: Re: [dpdk-dev]

Re: [dpdk-dev] [PATCH 0/6] fix port id type

On 10/13/2017 3:04 AM, Yang, Zhiyong wrote: > Hi Ferruh, > > The patch (ethdev: increase port_id range) tend to increase port id range in > ethdev, not eventdev, > Do you mean we should change eventdev port id ? > > Before V2, that patchset includes eventdev port id extension. > > but Bruce

Re: [dpdk-dev] [PATCH 0/6] fix port id type

Hi Ferruh, The patch (ethdev: increase port_id range) tend to increase port id range in ethdev, not eventdev, Do you mean we should change eventdev port id ? Before V2, that patchset includes eventdev port id extension. but Bruce reminded me about it in that patchset v2 thread and suggest to

[dpdk-dev] [PATCH v7 3/4] vhost: support to kick in secondary process

To support kick in secondary process, we propose callfd_pri and kickfd_pri to store the value in primary process; and by a new API, rte_vhost_set_vring_effective_fd(), we can set effective callfd and kickfd which can be used by secondary process. Note in this case, either primary process or the se

[dpdk-dev] [PATCH v7 4/4] net/vhost: support to run in the secondary process

Support to run vhost-pmd vdev in the secondary process. We obtain information, like memory regions, kickfd, callfd, through primary/secondary communication channel. And by invoking rte_vhost_set_vring_effective_fd, we can set the kickfd which can be recognized by the secondary process. Signed-off

[dpdk-dev] [PATCH v7 2/4] vhost: allocate virtio_net in memzone

Instead of allocate on the stack, change to allocate in memzone so that we can retrieve them in secondary processes. TODO: numa awareness. Signed-off-by: Jianfeng Tan --- lib/librte_vhost/socket.c | 2 ++ lib/librte_vhost/vhost.c | 34 -- lib/librte_vhost/vhost

[dpdk-dev] [PATCH v7 1/4] ethdev: support attach vdev in secondary process

When vdev driver requests an ethdev entry in secondary process, we will identify the correct entry in rte_eth_dev_data array and return the correct entry in the rte_eth_devices arrays. Signed-off-by: Jianfeng Tan --- lib/librte_ether/rte_ethdev_vdev.h | 26 +++--- 1 file chan

[dpdk-dev] [PATCH v7 0/4] move vdev into drivers/bus

v7: - Add notice in release note for API change, and bump library version of librte_cryptodev and librte_eal, as suggested by Thomas. v6: - Don't introduce the static log type for bug, instead we use dynamic log type for vdev, suggested by Shreyansh Jain. v4 & v5: - Fix issues of co

Re: [dpdk-dev] [PATCH v8 0/2] net/i40e: API to configure queue regions for RSS

Thank you. > -Original Message- > From: Yigit, Ferruh > Sent: Thursday, October 12, 2017 5:07 AM > To: Zhao1, Wei ; dev@dpdk.org > Subject: Re: [dpdk-dev] [PATCH v8 0/2] net/i40e: API to configure queue > regions for RSS > > On 10/11/2017 9:55 AM, Wei Zhao wrote: > > The patches mainly f

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

-Original Message- > Date: Fri, 13 Oct 2017 09:16:31 +0800 > From: Jia He > To: Jerin Jacob , "Ananyev, Konstantin" > > Cc: Olivier MATZ , "dev@dpdk.org" , > "jia...@hxt-semitech.com" , > "jie2@hxt-semitech.com" , > "bing.z...@hxt-semitech.com" > Subject: Re: [PATCH] ring: guaran

Re: [dpdk-dev] [PATCH v3 15/19] vhost: postpone rings addresses translation

Hi, Maxime After this commit, vhost/virtio loopback test will fail when use the CPU on socket 1. Error message like following during initialize: VHOST_CONFIG: vring kick idx:0 file:20 VHOST_CONFIG: reallocate vq from 0 to 1 node VHOST_CONFIG: reallocate dev from 0 to 1 node VHOST_CONFIG: (0) fai

Re: [dpdk-dev] [dpdk-stable] [PATCH] app/testpmd: fix runtime dead code

> -Original Message- > From: Yigit, Ferruh > Sent: Friday, October 13, 2017 3:28 AM > To: Jastrzebski, MichalX K ; Wu, Jingjing > > Cc: dev@dpdk.org; Jain, Deepak K ; Kulasek, > TomaszX ; Chilikin, Andrey > ; sta...@dpdk.org > Subject: Re: [dpdk-stable] [PATCH] app/testpmd: fix runtime d

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

Hi On 10/13/2017 9:02 AM, Jia He Wrote: Hi Jerin On 10/13/2017 1:23 AM, Jerin Jacob Wrote: -Original Message- Date: Thu, 12 Oct 2017 17:05:50 + [...] On the same lines, Jia He, jie2.liu, bing.zhao, Is this patch based on code review or do you saw this issue on any of the

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

On 10/13/2017 9:02 AM, Jia He Wrote: Hi Jerin On 10/13/2017 1:23 AM, Jerin Jacob Wrote: -Original Message- Date: Thu, 12 Oct 2017 17:05:50 + [...] On the same lines, Jia He, jie2.liu, bing.zhao, Is this patch based on code review or do you saw this issue on any of the arm

Re: [dpdk-dev] [PATCH v7 1/3] eal/x86: run-time dispatch over memcpy

Hi Sorry for the late reply. I took AL last 3 days. > -Original Message- > From: Thomas Monjalon [mailto:tho...@monjalon.net] > Sent: Tuesday, October 10, 2017 01:47 > To: Li, Xiaoyun > Cc: dev@dpdk.org; Ananyev, Konstantin ; > Richardson, Bruce ; Lu, Wenzhuo > ; Zhang, Helin > Subject:

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

Hi Jerin On 10/13/2017 1:23 AM, Jerin Jacob Wrote: -Original Message- Date: Thu, 12 Oct 2017 17:05:50 + From: "Ananyev, Konstantin" To: Olivier MATZ , Jia He CC: "dev@dpdk.org" , "jia...@hxt-semitech.com" , "jie2@hxt-semitech.com" , "bing.z...@hxt-semitech.com" , "jerin

Re: [dpdk-dev] [PATCH v7 0/3] run-time Linking support

Hi Sorry for the late reply. The original rte_memcpy.h is in include directory. So I just add them there. But you are right, I should move them to common/arch/x86/ directory. I will modify that in next version. Best Regards Xiaoyun Li > -Original Message- > From: Thomas Monjalon [mailt

Re: [dpdk-dev] [PATCH v7 5/6] igb_uio: use kernel functions for masking MSI-X

On 10/12/2017 10:15 PM, Stephen Hemminger wrote: > Are the symbols exported? No! They started to be exported in v4.5. I will update accordingly. Thanks, ferruh > > On Oct 12, 2017 13:58, "Ferruh Yigit" > wrote: > > On 10/12/2017 11:50 AM, Kavanagh, Mark B w

[dpdk-dev] [PATCH 4/4] eal/x86: implement arch-specific TSC freq query

From: Sergio Gonzalez Monroy First, try to use CPUID Time Stamp Counter and Nominal Core Crystal Clock Information Leaf to determine the tsc hz on platforms that supports it (does not require privileged user). If the CPUID leaf is not available, then try to determine the tsc hz by reading the MS

[dpdk-dev] [PATCH 0/4] improve tsc frequency calibration

v3 changes: - implement in .c files instead of exporting an inline arch function - rename arch function from rte_rdtsc_arch_hz to get_tsc_freq_arch - integrate x86 implementation in the series - fix private EAL include in PPC implementation (not tested) From: Gowrishankar Muthukrishnan Some arc

[dpdk-dev] [PATCH 2/4] eal/armv8: implement arch-specific TSC freq query

From: Jerin Jacob Use cntvct_el0 system register to get the system counter frequency. If the system is configured with RTE_ARM_EAL_RDTSC_USE_PMU then return 0(let the common code calibrate the tsc frequency). Signed-off-by: Jerin Jacob Acked-by: Jianbo Liu --- lib/librte_eal/common/arch/arm/

[dpdk-dev] [PATCH 3/4] eal/ppc64: implement arch-specific TSC freq query

From: Jerin Jacob In ppc_64, rte_rdtsc() returns timebase register value which increments at independent timebase frequency and hence not related to lcore cpu frequency to derive TSC hz. Hence, we stick with master lcore frequency. Signed-off-by: Jerin Jacob Signed-off-by: Gowrishankar Muthukri

[dpdk-dev] [PATCH 1/4] timer: honor arch-specific TSC frequency query

From: Jerin Jacob When calibrating the TSC frequency, first, probe the architecture specific function. If not available, use the existing calibrate scheme. Signed-off-by: Jerin Jacob Signed-off-by: Thomas Monjalon --- lib/librte_eal/bsdapp/eal/Makefile | 1 + lib/librte_eal/commo

Re: [dpdk-dev] Vendor specific sub-trees under next-net

13/10/2017 01:31, Ferruh Yigit: > Hi Thomas, et al > > Previously it has been mentioned [1] to have vendor specific driver > trees under next-net. > > And recently Mellanox agreed to have a Mellanox tree [2]. > > Intel also agrees to have next-net-intel, and Helin will be maintaining > it, thank

Re: [dpdk-dev] Vendor specific sub-trees under next-net

On 10/13/2017 12:29 AM, Ferruh Yigit wrote: > Hi Thomas, et al > > Previously it has been mentioned [1] to have vendor specific driver > trees under next-net. > > And recently Mellanox agreed to have a Mellanox tree [2]. > > Intel also agrees to have next-net-intel, and Helin will be maintaining

[dpdk-dev] Vendor specific sub-trees under next-net

Hi Thomas, et al Previously it has been mentioned [1] to have vendor specific driver trees under next-net. And recently Mellanox agreed to have a Mellanox tree [2]. Intel also agrees to have next-net-intel, and Helin will be maintaining it, thanks to Helin for volunteering. Other vendors with m

Re: [dpdk-dev] [PATCH v2 2/5] eal/ppc64: define architecture specific rdtsc hz

22/09/2017 10:25, Gowrishankar: > From: Jerin Jacob > > In ppc_64, rte_rdtsc() returns timebase register value which increments > at independent timebase frequency and hence not related to lcore cpu > frequency to derive TSC hz. Hence, we stick with master lcore frequency. > > CC: Chao Zhu > Si

Re: [dpdk-dev] [PATCH v7 5/6] igb_uio: use kernel functions for masking MSI-X

Are the symbols exported? On Oct 12, 2017 13:58, "Ferruh Yigit" wrote: > On 10/12/2017 11:50 AM, Kavanagh, Mark B wrote: > > Hi, > > > > This commit renders igb_uio unusable on Fedora 20, kernel version > 3.19.8-100.fc20.x86_64. > > > > During the build (make -j 20), a warning is issued for igb_

Re: [dpdk-dev] [PATCH v7 5/6] igb_uio: use kernel functions for masking MSI-X

On 10/12/2017 11:50 AM, Kavanagh, Mark B wrote: > Hi, > > This commit renders igb_uio unusable on Fedora 20, kernel version > 3.19.8-100.fc20.x86_64. > > During the build (make -j 20), a warning is issued for igb_uio regarding a > missing symbol (pci_msi_unmask): > WARNING: "pci_msi_unmas

Re: [dpdk-dev] [PATCH v7 1/3] app/test-pmd: add CLI for TM capability and stats

Jingjing, please could you review this series? Thanks 11/10/2017 11:26, Jasvinder Singh: > Add following CLIs to testpmd application for device traffic management; > - commands to display TM capability information. > (per port, per hierarchy level and per hierarchy node) > - command to display h

Re: [dpdk-dev] [PATCH 0/4] meter: add meter configuration profile api

> Cristian Dumitrescu (4): > meter: add meter configuration profile > test: meter autotest update > qos_meter: accommodate meter api changes > deprecation: removed the librte_meter notice > > doc/guides/rel_notes/deprecation.rst | 3 - > examples/qos_meter/main.c | 36 ++

[dpdk-dev] Support for Intel 82579LM (8086:1502)

Hello, I have a computer I’m using for testing DPDK that has an Intel 82579LM. Currently this NIC isn’t supported in DPDK, but I believe it is similar to others that are (such as the 82574L, which I am using successfully on another computer). How much work would be required in getting this up an

Re: [dpdk-dev] [PATCH v8 0/6] Reproducible build

> Luca Boccassi (6): > mk: sort list of shared objects in linker script > mk: sort list of files in examples.dox > mk: sort headers before wildcard inclusion > mk: sort source files before passing them to the compiler > mk: sort object files when building deps lists > mk: install symlin

Re: [dpdk-dev] [PATCH v3 1/3] eal: move bitmap from lib sched

> > From: Pavan Bhagavatula > > > > The librte_sched uses rte_bitmap to manage large arrays of bits in an > > optimized method so, moving it to eal/common would allow other libraries > > and applications to use it. > > > > Signed-off-by: Pavan Nikhilesh > > Thanks very much, Pavan! > > Series

Re: [dpdk-dev] [PATCH 0/6] fix port id type

On 10/12/2017 7:03 PM, Ferruh Yigit wrote: > On 10/12/2017 10:32 AM, Zhiyong Yang wrote: >> Now ethdev port_id has been increased to uint16_t from uint8_t in the below >> patch >> http://www.dpdk.org/dev/patchwork/patch/29418/ >> >> but >> 1. Some features which are applied were still developed ba

Re: [dpdk-dev] [PATCH 0/2] enable 4KB + VFIO-PCI

On 10/11/2017 11:33 AM, Jianfeng Tan wrote: > Patch 1: Use VA as IOVA if IOVA mode is enabled. > Patch 2: Enable IOVA mode for the PMDs for Intel NICs. > > How to test: > > $ (bind nic to vfio-pci) > $ testpmd -c 0x3 -n 4 -m 2048 --no-huge -- -i --no-numa > > Jianfeng Tan (2): > eal: honor IOV

Re: [dpdk-dev] [PATCH v2] ethdev: fix xstats retrieve by id API

On 10/12/2017 2:31 PM, Lee Daly wrote: > From: Lee > > Fix xstats functions, rte_eth_xstats_get_names_by_id() > and rte_eth_xstats_get_by_id(), in current implementation > ethdev level reads all xstat values and filters out > the ones requested by the application. This behavior doesn't > benefit

Re: [dpdk-dev] [dpdk-stable] [PATCH] app/testpmd: fix runtime dead code

On 9/29/2017 1:56 PM, Michal Jastrzebski wrote: > From: Tomasz Kulasek > > Coverity reports DEADCODE, as assumes that RTE_LIBRTE_I40E_PMD > is defined and function compiles entirely. > The fix is about to place printf function into #else branch > of conditional compilation, since it is known at

Re: [dpdk-dev] [PATCH v2 00/29] net/mlx4: restore PMD functionality

On 10/12/2017 1:19 PM, Adrien Mazarguil wrote: > This series restores all the control path functionality removed in prior > series "net/mlx4: trim and refactor entire PMD", including: > > - Promiscuous mode. > - All multicast mode. > - MAC address configuration. > - Support for multiple simultaneo

[dpdk-dev] [PATCH] eventdev: use new event enqueue in Rx adapter

When event Rx adapter uses software based packet transfer, event op is always RTE_EVENT_OP_NEW. By using rte_event_enqueue_new_burst API the underlying event devices driver can use it as a hint to optimize enqueue operation if possible. Signed-off-by: Pavan Nikhilesh --- lib/librte_eventdev/rte_

Re: [dpdk-dev] [PATCH 0/6] fix port id type

On 10/12/2017 10:32 AM, Zhiyong Yang wrote: > Now ethdev port_id has been increased to uint16_t from uint8_t in the below > patch > http://www.dpdk.org/dev/patchwork/patch/29418/ > > but > 1. Some features which are applied were still developed based on older version > uint8_t port_id. > 2. Some

[dpdk-dev] [PATCH 2/3] net/bnxt: remove redundant code parsing pool_map

This patch removes some redundant code from bnxt_mq_rx_configure(). Fixes: 8056afdf97d7 ("net/bnxt: fix the association of a MACVLAN per VNIC) Signed-off-by: Ajit Khaparde --- drivers/net/bnxt/bnxt_rxq.c | 11 --- 1 file changed, 11 deletions(-) diff --git a/drivers/net/bnxt/bnxt_rxq.c

[dpdk-dev] [PATCH 1/3] net/bnxt: handle Rx multi queue creation properly

This patch simplifies logic for RSS queue creation. 1) Do not hardcode number of VNIC pools in case of RSS 2) Log a message if Number of queues is > RTE_ETHDEV_QUEUE_STAT_CNTRS 3) Move the check for allocation of l2_filter inside the for loop. Fixes: 6133f207970c ("net/bnxt: add Rx queue create/de

[dpdk-dev] [PATCH 0/3] patchset for bnxt PMD

Patchset against dpdk-next Apart from couple of fixes, this set updates the HWRM version to 1.8.2. The patch to update the HWRM is based on a file which is autogenerated and most of the changes in the actual PMD are macro name changes. This patch will throw a bunch of checkpatch warnings, but they

Re: [dpdk-dev] [dpdk-stable] [PATCH] net/i40e: fix unexpected mbuf free in vPMD

On 10/10/2017 2:22 PM, Qi Zhang wrote: > vPMD tx does not set sw_ring's mbuf to NULL after free it. > So to prevent same mbuf be free again, we need more carefully > check in i40e_tx_queue_release_mbufs. > > Fixes: b4669bb95038 ("i40e: add vector Tx") > Cc: sta...@dpdk.org > > Signed-off-by: Qi Z

[dpdk-dev] [PATCH v2] net/e1000: correct VLAN tag byte order for i35x LB packets

When copying VLAN tags from the RX descriptor to the vlan_tci field in the mbuf header, igb_rxtx.c:eth_igb_recv_pkts() and eth_igb_recv_scattered_pkts() both assume that the VLAN tag is always little endian. While i350, i354 and /i350vf VLAN non-loopback packets are stored little endian, VLAN tag

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

-Original Message- > Date: Thu, 12 Oct 2017 17:05:50 + > From: "Ananyev, Konstantin" > To: Olivier MATZ , Jia He > CC: "dev@dpdk.org" , "jia...@hxt-semitech.com" > , "jie2@hxt-semitech.com" > , "bing.z...@hxt-semitech.com" > , "jerin.ja...@caviumnetworks.com" > > Subject: RE:

Re: [dpdk-dev] [PATCH] net/i40e: fix flexible payload configuration

On 10/12/2017 11:13 AM, Xing, Beilei wrote: > >> -Original Message- >> From: Chilikin, Andrey >> Sent: Saturday, October 7, 2017 2:11 AM >> To: dev@dpdk.org >> Cc: Xing, Beilei ; Wu, Jingjing >> ; Chilikin, Andrey >> Subject: [PATCH] net/i40e: fix flexible payload configuration >> >> Remo

Re: [dpdk-dev] [pull-request] next-pipeline 17.11 pre-rc1

Hi, 11/10/2017 16:40, Cristian Dumitrescu: > The following changes since commit d65b3b1668f2037745407c3909c43d85a18692a6: > > vhost: fix false-positive warning from clang 5 (2017-10-11 13:56:34 +0200) > > are available in the git repository at: > > http://dpdk.org/git/next/dpdk-next-pipelin

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

Hi guys, > -Original Message- > From: Olivier MATZ [mailto:olivier.m...@6wind.com] > Sent: Thursday, October 12, 2017 4:54 PM > To: Jia He > Cc: dev@dpdk.org; jia...@hxt-semitech.com; jie2@hxt-semitech.com; > bing.z...@hxt-semitech.com; Ananyev, Konstantin > ; jerin.ja...@caviumnetwo

[dpdk-dev] [RFC] Compression API in DPDK

With the vast amounts of data being transported around networks and stored in storage systems, reducing data size is becoming ever more important. There are both software libraries and hardware devices available that provide compression, but no common API. This RFC proposes a compression API for

Re: [dpdk-dev] [PATCH v7 5/6] igb_uio: use kernel functions for masking MSI-X

On 10/9/2017 10:56 PM, Patrick MacArthur wrote: > Hi, Markus, > > This commit appears to cause a regression on CentOS 7.4 with the in-box > Linux 3.10.0-639-2.2.el7.x86_64 kernel. Although the kernel module > appears to build correctly, when I attempt to load the module with > insmod, it fails

Re: [dpdk-dev] [PATCH v1 0/7] Flow API helpers enhancements

On 10/12/2017 1:53 PM, Adrien Mazarguil wrote: > On Wed, Oct 11, 2017 at 07:07:29PM +0100, Ferruh Yigit wrote: >> On 10/11/2017 10:57 AM, Adrien Mazarguil wrote: >>> On Tue, Oct 10, 2017 at 07:05:30PM +0100, Ferruh Yigit wrote: > After above said, API changes one week before integration deadl

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

On Thu, 12 Oct 2017 17:53:51 +0200 Olivier MATZ wrote: > Hi, > > On Tue, Oct 10, 2017 at 05:56:36PM +0800, Jia He wrote: > > Before this patch: > > In __rte_ring_move_cons_head() > > ... > > do { > > /* Restore n as it may change every loop */ > > n = max;

Re: [dpdk-dev] [PATCH v5 1/4] ethdev: add support for raw flow type for flow director

Hi Kirill, On Thu, Oct 12, 2017 at 11:41:50AM +, Rybalchenko, Kirill wrote: > Hi Thomas, > > The reason this feature is needed is to be able to program custom flow types > using a template packet rather than building up a C struct to define the > protocol. This means that users don't have t

[dpdk-dev] [PATCH V3] doc: minor fixes for GSO prog_guide

Correct two minor issues in the GSO programmer's guide: - a note is rendered incorrectly in the middle of an unordered list; this results in the remainder of the list appearing inside the note. Correct indentation of the note to resolve same. - two minor visual artifacts are present in the 'thr

[dpdk-dev] [PATCH 3/3] mk: do not generate LDLIBS from directory dependencies

The list of libraries in LDLIBS was generated from the DEPDIRS-xyz variable. This is valid when the subdirectory name match the library name, but it's not always the case, especially for PMDs. The patches removes this feature and explicitly adds the proper libraries in LDLIBS. Some DEPDIRS-xyz va

[dpdk-dev] [PATCH 2/3] mempool/octeontx: fix dependency

The syntax "DEPDIRS-y = ..." is deprecated since commit feb9f680cd2c ("mk: optimize directory dependencies"). Replace it with the new style. Fixes: 8700239f7767 ("mempool/octeontx: add build and log infrastructure") Cc: sta...@dpdk.org Signed-off-by: Olivier Matz --- drivers/mempool/Makefile

[dpdk-dev] [PATCH 1/3] crypto/dpaa2_sec: remove uneffective dependency

The syntax "DEPDIRS-y = ..." is deprecated since commit feb9f680cd2c ("mk: optimize directory dependencies"). Remove this uneffective lines. The effective dependencies are already declared in drivers/crypto/Makefile, so there is nothing to add. Signed-off-by: Olivier Matz --- drivers/crypto/dp

[dpdk-dev] [PATCH 0/3] mk: fix LDLIBS

Some libraries in LDLIBS are currently generated from the DEPDIRS-$(dirname) variables in the Makefiles. This is valid when the subdirectory name match the library name, but it's not always the case, especially for PMDs. This series fixes some invalid dependency declarations first, and then remove

[dpdk-dev] [PATCH V2] doc: minor fixes for GSO prog_guide

Correct two minor issues in the GSO programmer's guide: - a note is rendered incorrectly in the middle of an unordered list; this results in the remainder of the list appearing inside the note. Correct indentation of the note to resolve same. - two minor visual artifacts are present in the 'thr

Re: [dpdk-dev] [PATCH] doc: minor fixes for GSO prog_guide

>From: Mcnamara, John >Sent: Thursday, October 12, 2017 4:29 PM >To: Kavanagh, Mark B ; dev@dpdk.org >Cc: Hu, Jiayu ; Tan, Jianfeng ; >Ananyev, Konstantin ; Yigit, Ferruh >; tho...@monjalon.net; Kavanagh, Mark B > >Subject: RE: [dpdk-dev] [PATCH] doc: minor fixes for GSO prog_guide > > > >> -Or

Re: [dpdk-dev] [PATCH] ring: guarantee ordering of cons/prod loading when doing enqueue/dequeue

Hi, On Tue, Oct 10, 2017 at 05:56:36PM +0800, Jia He wrote: > Before this patch: > In __rte_ring_move_cons_head() > ... > do { > /* Restore n as it may change every loop */ > n = max; > > *old_head = r->cons.head;//1st load >

[dpdk-dev] [PATCH 1/2] vhost: fix deadlock on IOTLB miss

An optimization was done to only take the iotlb cache lock once per packet burst instead of once per IOVA translation. With this, IOTLB miss requests are sent to Qemu with the lock held, which can cause a deadlock if the socket buffer is full, and if Qemu is waiting for an IOTLB update to be done.

[dpdk-dev] [PATCH 2/2] vhost: fix IOTLB on NUMA realloc

In case of NUMA reallocation, virtqueue's iotlb list is broken, has its head changes but first iotlb entry in the list still points to the previous head pointer. Also, in case of reallocation, we want the IOTLB cache mempool to be on the new socket. This patch perform a full re-init of the IOTLB

[dpdk-dev] [PATCH 0/2] vhost: IOTLB fixes for -rc1

These two patches fixes issues faced when running the VM on a different socket than DPDK. In this case, the numa_realloc() function is called to reallocate the virtqueue and the virtio-net device structs on the VM's socket. The problem is that doing so corrupts the IOTLB cache list, as the list h

[dpdk-dev] Need help to run test_pmd via sr-iov setup

Need help to run test_pmd via sr-iov setup my steps:1. turn on intel vt in bios2. set "intel_iommu=on" in kernel boot parameter3. update host grub and reboot4. create VF:    echo 1 > /sys/class/net/eth17/device sriov_numvfs    echo 1 > /sys/class/net/eth18/device sriov_numvfs    cat /sys/cla

Re: [dpdk-dev] [PATCH] doc: minor fixes for GSO prog_guide

> -Original Message- > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Mark Kavanagh > Sent: Thursday, October 12, 2017 2:46 PM > To: dev@dpdk.org > Cc: Hu, Jiayu ; Tan, Jianfeng > ; Ananyev, Konstantin > ; Yigit, Ferruh ; > tho...@monjalon.net; Kavanagh, Mark B > Subject: [dpdk-dev

Re: [dpdk-dev] [pull-request] next-crypto 17.11 rc1

12/10/2017 06:38, Pablo de Lara: > http://dpdk.org/git/next/dpdk-next-crypto Pulled, thanks

Re: [dpdk-dev] [PATCH V3 4/5] doc: ethdev traffic metering and policing api

Reviewed-by: John McNamara

Re: [dpdk-dev] [PATCH V3 4/5] doc: ethdev traffic metering and policing api

> -Original Message- > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Cristian Dumitrescu > Sent: Friday, October 6, 2017 3:46 PM > To: dev@dpdk.org > Cc: tho...@monjalon.net; adrien.mazarg...@6wind.com; Wu, Jingjing > ; hemant.agra...@nxp.com; > jerin.ja...@caviumnetworks.com; Sing

Re: [dpdk-dev] [PATCH] eal: bus scan and probe never fail

Hello Aaron, On Thursday 12 October 2017 06:50 PM, Aaron Conole wrote: Shreyansh Jain writes: Hello Aaron, On Tuesday 10 October 2017 09:30 PM, Aaron Conole wrote: Shreyansh Jain writes: Hello Don, [snip] [snip] I am assuming that that aim of this is to have a way so that appl

Re: [dpdk-dev] [PATCH v1 1/7] ethdev: expose flow API error helper

On Wed, Oct 11, 2017 at 03:05:47PM -0400, Aaron Conole wrote: > Adrien Mazarguil writes: > > > On Wed, Oct 11, 2017 at 11:23:12AM +0200, Thomas Monjalon wrote: > >> 05/10/2017 11:49, Adrien Mazarguil: > >> > rte_flow_error_set() is a convenient helper to initialize error objects. > >> > > >> > S

Re: [dpdk-dev] [PATCH v4 02/16] net/mrvl: add mrvl net pmd driver skeleton

Le 12/10/2017 à 14:14, Jacek Siuda a écrit : What we can do, is to add a run-time driver option and oblige user (in documentation) to set the option only for the first mrvl vdev. If the driver doesn't see the option set, it  won't initialize DMA. This, plus meaningful error handling/logs shou

[dpdk-dev] [PATCH] doc: minor fixes for GSO prog_guide

Correct two minor issues in the GSO programmer's guide: - a note is rendered incorrectly in the middle of an unordered list; this results in the remainder of the list appearing inside the note. Move note to end of the list to resolve same. - two minor visual artifacts are present in the 'three-

Re: [dpdk-dev] [PATCH v3 08/12] doc: add details of rte_flow security actions

> -Original Message- > From: Akhil Goyal [mailto:akhil.go...@nxp.com] > Sent: Friday, October 6, 2017 7:12 PM > To: dev@dpdk.org > Cc: Doherty, Declan ; De Lara Guarch, Pablo > ; hemant.agra...@nxp.com; Nicolau, Radu > ; bor...@mellanox.com; avia...@mellanox.com; > tho...@monjalon.net; sa

  1   2   3   >