[dpdk-dev] [PATCH 1/2] virtio: fixed segmentation fault in queue_release

2015-10-20 11:44, Stephen Hemminger: > On Tue, 20 Oct 2015 16:37:41 +0100 > Bernard Iremonger wrote: > > > if input parameter vq is NULL, hw = vq->hw, causes a segmentation fault. > > > > Signed-off-by: Bernard Iremonger > > This could only happen for the case of the control vq. > Others are g

[dpdk-dev] [PATCH v3 7/7] virtio: pick simple rx/tx func

simple rx/tx func is enabled when user specifies single segment and no offload support. merge-able should be disabled to use simple rxtx. Signed-off-by: Huawei Xie --- drivers/net/virtio/virtio_rxtx.c | 12 1 file changed, 12 insertions(+) diff --git a/drivers/net/virtio/virtio_rx

[dpdk-dev] [PATCH v3 6/7] virtio: simple tx routine

Changes in v3: - Remove return at the end of void function - Remove always_inline attribute for virtio_xmit_cleanup bulk free of mbufs when clean used ring. shift operation of idx could be saved if vq_free_cnt means free slots rather than free descriptors. TODO: rearrange vq data structure, pack

[dpdk-dev] [PATCH v3 5/7] virtio: virtio vec rx

With fixed avail ring, we don't need to get desc idx from avail ring. virtio driver only has to deal with desc ring. This patch uses vector instruction to accelerate processing desc ring. Signed-off-by: Huawei Xie --- drivers/net/virtio/virtio_ethdev.h | 2 + drivers/net/virtio/virtio_rxt

[dpdk-dev] [PATCH v3 4/7] virtio: fill RX avail ring with blank mbufs

fill avail ring with blank mbufs in virtio_dev_vring_start Signed-off-by: Huawei Xie --- drivers/net/virtio/Makefile | 2 +- drivers/net/virtio/virtio_rxtx.c| 6 ++- drivers/net/virtio/virtio_rxtx.h| 3 ++ drivers/net/virtio/virtio_rxtx_simple.c | 84 ++

[dpdk-dev] [PATCH v3 3/7] virtio: rx/tx ring layout optimization

In DPDK based switching enviroment, mostly vhost runs on a dedicated core while virtio processing in guest VMs runs on other different cores. Take RX for example, with generic implementation, for each guest buffer, a) virtio driver allocates a descriptor from free descriptor list b) modify the entr

[dpdk-dev] [PATCH v3 2/7] virtio: add software rx ring, fake_buf into virtqueue

Changes in v3: - Remove unnecessary NULL test for rte_free - Remove unnecessary assign of local var vq after free Add software RX ring in virtqueue. Add fake_mbuf in virtqueue for wraparound processing. Use global simple_rxtx to indicate whether simple rxtx is enabled Signed-off-by: Huawei Xie -

[dpdk-dev] [PATCH v3 1/7] virtio: add virtio_rxtx.h header file

Would move all rx/tx related declarations into this header file in future. Add RTE_VIRTIO_PMD_MAX_BURST. Signed-off-by: Huawei Xie --- drivers/net/virtio/virtio_ethdev.c | 1 + drivers/net/virtio/virtio_rxtx.c | 1 + drivers/net/virtio/virtio_rxtx.h | 34 ++

[dpdk-dev] [PATCH v3 0/7] virtio ring layout optimization and simple rx/tx processing

Changes in v2: - Remove the configure macro - Enable simple R/TX processing when user specifies simple txq flags - Reword some comments and commit messages Changes in v3: - Remove unnecessary NULL test for rte_free - Remove unnecessary assign of local var after free - Remove return at the end of v

[dpdk-dev] [PATCH] eal: move interrupt init after device init

2015-10-14 13:32, David Marchand: > On Wed, Oct 14, 2015 at 11:50 AM, David Marchand > wrote: > > > Fixes: "eal: set iopl only when needed" (565b85d) > > > > Forgot the release note. > If this current patch is ok for all, I will resubmit with an update to > doc/guides/rel_notes/release_2_2.rst.

[dpdk-dev] [PATCH 1/3] eal: new interrupt handler type

On Mon, Oct 5, 2015 at 7:56 PM, Adrien Mazarguil wrote: > From: Nelio Laranjeiro > > Add RTE_INTR_HANDLE_EXT handler type for PMDs that do not support VFIO or > UIO. Those are expected to manage the file descriptor themselves. > > Signed-off-by: Nelio Laranjeiro > Maybe a title like "eal: all

[dpdk-dev] [PATCH 1/3] mlx4: fix missing ol_flags in scattered RX function

2015-10-05 19:50, Adrien Mazarguil: > They were dropped by mistake in the commit below. > > Fixes: ab351fe1c95c ("mbuf: remove packet type from offload flags") > > Signed-off-by: Adrien Mazarguil Series applied, thanks

[dpdk-dev] [PATCH v7 3/8] vhost: vring queue setup for multiple queue support

On Wed, 21 Oct 2015 11:48:09 +0800 Yuanhan Liu wrote: > struct virtio_net { > - struct vhost_virtqueue *virtqueue[VIRTIO_QNUM];/**< Contains > all virtqueue information. */ > + struct vhost_virtqueue *virtqueue[VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX]; > /**< Contains all virtqueu

[dpdk-dev] [PATCH v7 4/8] vhost: rxtx: use queue id instead of constant ring index

On Wed, 21 Oct 2015 11:48:10 +0800 Yuanhan Liu wrote: > > +static inline int __attribute__((always_inline)) > +is_valid_virt_queue_idx(uint32_t virtq_idx, int is_tx, uint32_t max_qp_idx) > +{ > + if ((is_tx ^ (virtq_idx & 0x1)) || > + (virtq_idx >= max_qp_idx * VIRTIO_QNUM)) > +

[dpdk-dev] DPDK patch backlog

Patch backlog is not getting better, now at 486. How can we break this logjam? Do I need to make a new "ready for merge" tree?

[dpdk-dev] [PATCH 2/2] enic: fix hash creation when not using first numa node

> >If dpdk is run with memory only available on socket != 0, then hash > >creation will fail and flow director feature won't be available. > >Fix this by asking for allocation on caller socket. > > > >Signed-off-by: David Marchand > > Acked by: Sujith Sankar Series applied, thanks.

[dpdk-dev] [PATCH v3 2/2] doc: update release note for igb TSO support

Signed-off-by: Wang Xiao W --- doc/guides/rel_notes/release_2_2.rst | 4 1 file changed, 4 insertions(+) diff --git a/doc/guides/rel_notes/release_2_2.rst b/doc/guides/rel_notes/release_2_2.rst index 5687676..74ebd49 100644 --- a/doc/guides/rel_notes/release_2_2.rst +++ b/doc/guides/rel_no

[dpdk-dev] [PATCH v3 1/2] e1000: enable igb TSO support

This patch enables igb TSO feature, the feature works on both PF and VF. The TCP segmentation offload needs to write the offload related information into the advanced context descriptors, which is similar to checksum offload. Signed-off-by: Wang Xiao W --- drivers/net/e1000/igb_ethdev.c | 6 +-

[dpdk-dev] [PATCH v3 0/2] e1000: enable igb TSO support

v3: * Removed the "unlikely" in check_tso_para function, for there was no obvious performance difference, let the branch predictor do the job. v2: * Reworded the old comment about union igb_vlan_macip which was no more used. * Corrected typo in line "There're some limitaions in hardware for TC

[dpdk-dev] [PATCH v2 0/6] cxgbe: Optimize tx/rx for 40GbE and add Jumbo Frame support for CXGBE PMD

2015-10-08 19:16, Rahul Lakkireddy: > This series of patches improve forwarding performance for Chelsio T5 40GbE > cards and add Jumbo Frame support for cxgbe pmd. Also update documentation > and release notes. Applied, thanks. What are the performance numbers with these patches?

[dpdk-dev] [PATCH 5/6] doc: Update BNX2X PMD documentation

This patch can be avoided by updating the documentation with each code change atomically.

[dpdk-dev] [PATCH 4/6] config: Enable BNX2X driver build by default

2015-10-08 09:54, Rasesh Mody: > From: Harish Patil > > Signed-off-by: Harish Patil > Signed-off-by: Rasesh Mody You cannot enable bnx2x without gracefully handle miss of zlib header.

[dpdk-dev] [PATCH 2/5] bnx2x: Fix x86_64-native-linuxapp-clang build error

2015-09-11 17:42, Rasesh Mody: > #define SHMEM_EEE_ADV_STATUS_MASK 0x00f0 > #define SHMEM_EEE_100M_ADV (1<<0) > - #define SHMEM_EEE_1G_ADV (1<<1) > + #define SHMEM_EEE_1G_ADV (1U<<1) > #define SHMEM_E

[dpdk-dev] [PATCH 1/5] bnx2x: SR-IOV VF driver changes to support OOB PF driver

2015-09-11 17:42, Rasesh Mody: > From: Harish Patil > > Signed-off-by: Harish Patil Please explain your change a bit more.

[dpdk-dev] [PATCH] rte_eth_af_packet: check return value from sendto call

2015-08-27 13:50, John W. Linville: > Coverity CID # 13200 > > If sendto fails, the packets will not get transmitted. Return 0 as > the number of packets transmitted. > > Signed-off-by: John W. Linville Applied, thanks

[dpdk-dev] [PATCH v2] mem: command line option to delete hugepage backing files

When an application using huge-pages crash or exists, the hugetlbfs backing files are not cleaned up. This is a patch to clean those files. There are multi-process DPDK applications that may be benefited by those backing files. Therefore, I have made that configurable so that the application that d

[dpdk-dev] [dpdk-dev, PATCHv6 1/6] ethdev: enhance rte_eth_(tx|rx)q_info struct

2015-10-20 09:36, Ananyev, Konstantin: > As I can see your patch series should be applied on top of mine: > [PATCHv5 0/8] ethdev: add new API to retrieve RX/TX queue information > Which is not yet in the dpdk.org mainline. > I believe that is ok, but then you shouldn't replace previous version with

[dpdk-dev] [PATCH] vhost: call write barrier before used index update

Descriptors that have been put into the used vring must be observable by guest earlier than the new used index value. Although compiler barrier serves well for Intel architectue here, the proper cross-platform solution is to use write barrier before the used index is updated. Signed-off-by: Nikita

[dpdk-dev] [PATCH v2] mem: command line option to delete hugepage backing files

When an application using huge-pages crash or exists, the hugetlbfs backing files are not cleaned up. This is a patch to clean those files. There are multi-process DPDK applications that may be benefited by those backing files. Therefore, I have made that configurable so that the application tha

[dpdk-dev] [dpdk-dev, PATCHv6 1/6] ethdev: enhance rte_eth_(tx|rx)q_info struct

> Yep, similar thought here: > In the for Intel HW implementations: > qinfo->used_desc = ixgbe_dev_rx_queue_count(dev, queue_id); > It seems a bit redundant, as if user wants to know HW state it can call > rte_eth_rx_queue_count() directly. > From other side: rte_eth_rx_queue_count() is quite heav

[dpdk-dev] [PATCH 2/2] rel_notes: update for fix for virtio segmentation fault

Signed-off-by: Bernard Iremonger --- doc/guides/rel_notes/release_2_2.rst | 3 +++ 1 file changed, 3 insertions(+) diff --git a/doc/guides/rel_notes/release_2_2.rst b/doc/guides/rel_notes/release_2_2.rst index 5687676..7e8bd4f 100644 --- a/doc/guides/rel_notes/release_2_2.rst +++ b/doc/guides/r

[dpdk-dev] [PATCH 1/2] virtio: fixed segmentation fault in queue_release

if input parameter vq is NULL, hw = vq->hw, causes a segmentation fault. Signed-off-by: Bernard Iremonger --- drivers/net/virtio/virtio_ethdev.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/virtio/virtio_ethdev.c b/drivers/net/virtio/virtio_ethdev.c index 46

[dpdk-dev] [PATCH v3 3/3] examples/ip_pipeline: add mp/mc and frag/ras swq

Add integrated MP/MC and fragmentation/reassembly support to SWQs Signed-off-by: Piotr Azarewicz --- examples/ip_pipeline/app.h | 14 +++ examples/ip_pipeline/config_check.c | 45 +++- examples/ip_pipeline/config_parse.c | 195 +-- examples/ip_pi

[dpdk-dev] [PATCH v3 2/3] port: fix ras ring ports

Bug fixes for ring ports with IPv4/IPv6 reassembly support. Previous implementation can't work properly due to incorrect choosing process function. Also, assuming that, when processing ip packet, ip header is know we can set l3_len parameter here. Fixes: 50f54a84dfb7 ("port: add IPv6 reassembly po

[dpdk-dev] [PATCH v3 1/3] port: add mp/mc ring ports

ring_multi_reader input port (on top of multi consumer rte_ring) ring_multi_writer output port (on top of multi producer rte_ring) Signed-off-by: Piotr Azarewicz --- lib/librte_port/rte_port_ring.c | 311 +++--- lib/librte_port/rte_port_ring.h | 35 +++-

[dpdk-dev] [PATCH v3 0/3] ip_pipeline: add MP/MC and frag/ras support to SWQs

This patch set enhancement ip_pipeline application: - librte_port: add support for multi-producer/multi-consumer ring ports - librte_port: bug fixes for ring ports with IPv4/IPv6 reassembly support - ip_pipeline application: integrate MP/MC and fragmentation/reassembly support to SWQs v2 changes:

[dpdk-dev] [PATCH v2 0/4] vhost: Fix virtio-net on VHOST_USER_RESET_OWNER

Acked-by: Yuanhan Liu Thanks. --yliu On Mon, Oct 19, 2015 at 11:44:25AM +0200, Jerome Jutteau wrote: > Hi, > > I have a bug when Qemu with two vhost interfaces gently stops (SIGINT). > When stopping, it sends two RESET_OWNER for each interface: > - Before stopping, we have two interfac

[dpdk-dev] [PATCH] mem: Command line option to delete hugepage backing files

Hi Shesha, A few format comments: - Spacing/tabbing is all wrong for some reason, need to be fixed. - Signed-off is missing, see 'Contribute by sending patches' in http://dpdk.org/dev. - No capital letters in the commit title - there is a patch submitted with more detailed info regarding patch f

[dpdk-dev] [PATCH v6 06/13] virtio: read virtio_net_config correctly

On Tue, Oct 20, 2015 at 09:27:37AM +0200, Thomas Monjalon wrote: > 2015-10-20 15:23, Yuanhan Liu: > > On Tue, Oct 20, 2015 at 09:07:49AM +0200, Thomas Monjalon wrote: > > > 2015-10-09 13:46, Yuanhan Liu: > > > > From: Changchun Ouyang > > > > > > > > The old code adjusts the config bytes we want

[dpdk-dev] [PATCH v2 2/7] virtio: add software rx ring, fake_buf into virtqueue

On 10/19/2015 1:07 PM, Xie, Huawei wrote: > On 10/19/2015 12:20 PM, Stephen Hemminger wrote: > > On Sun, 18 Oct 2015 14:28:59 +0800 > Huawei Xie wrote: > > > > + if (vq->sw_ring) > + rte_free(vq->sw_ring); > + > > > > Do not need

[dpdk-dev] [PATCH] vhost: call write barrier before used index update

On Tue, Oct 20, 2015 at 05:07:46PM +0300, Nikita Kalyazin wrote: > Descriptors that have been put into the used vring must be observable by > guest earlier than the new used index value. > Although compiler barrier serves well for Intel architectue here, the > proper cross-platform solution is to u

[dpdk-dev] [PATCH v6 06/13] virtio: read virtio_net_config correctly

On Tue, Oct 20, 2015 at 09:07:49AM +0200, Thomas Monjalon wrote: > 2015-10-09 13:46, Yuanhan Liu: > > From: Changchun Ouyang > > > > The old code adjusts the config bytes we want to read depending on > > what kind of features we have, but we later cast the entire buf we > > read with "struct virt

[dpdk-dev] [PATCH] rte_ether: clarify rte_eth_set_queue_rate_limit tx_rate parameter

The tx_rate unit is Mbps. Gleaned from the ixgbe implementation, the 82599 datasheet and the use in test-pmd. Signed-off-by: Simon Kagstrom --- lib/librte_ether/rte_ethdev.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_

[dpdk-dev] [PATCH v6 05/13] vhost-user: handle VHOST_USER_RESET_OWNER correctly

On Tue, Oct 20, 2015 at 09:03:48AM +0200, Thomas Monjalon wrote: > 2015-10-09 13:46, Yuanhan Liu: > > Destroy corresponding device when a VHOST_USER_RESET_OWNER message is > > received, otherwise, the vhost-switch would still try to access vq > > of that device, which results to SIGSEG fault, and l

[dpdk-dev] When I run test-pmd, most of received packets(loop-backed packet) have RX-error.

Dear DPDK experts. Thank you very much for your best great efforts and precious answers. When I run test-pmd, most of received packets are RX-error. The computer has two 10GbE ports Intel NIC and the two ports are loop-backed each other. The result shows that the loop-backed packets have rx-e

[dpdk-dev] [PATCH v6 04/13] vhost: rxtx: prepare work for multiple queue support

On Tue, Oct 20, 2015 at 08:57:52AM +0200, Thomas Monjalon wrote: > 2015-10-09 13:46, Yuanhan Liu: > > From: Changchun Ouyang > > > > Do not use VIRTIO_RXQ or VIRTIO_TXQ anymore; use the queue_id, > > instead, which will be set to a proper value for a specific queue > > when we have multiple queue

[dpdk-dev] [PATCH v4 5/5] doc: modify release notes and deprecation notice for table and pipeline

From: Marcin Kerlin The release notes is updated and the deprecation announce is removed. Signed-off-by: Maciej Gajdzica Signed-off-by: Marcin Kerlin Acked-by: Cristian Dumitrescu --- doc/guides/rel_notes/deprecation.rst | 3 --- doc/guides/rel_notes/release_2_2.rst | 2 +- 2 files changed,

[dpdk-dev] [PATCH v4 4/5] ip_pipline: added cli commands for bulk add/delete to firewall pipeline

From: Marcin Kerlin Added two new cli commands to firewall pipeline. Commands bulk add and bulk delete takes as argument a file with rules to add/delete. The file is parsed, and then rules are passed to backend functions which add/delete records from pipeline tables. Signed-off-by: Maciej Gajdzi

[dpdk-dev] [PATCH v4 3/5] test_table: added check for bulk add/delete to acl table unit test

From: Marcin Kerlin Added to acl table unit test check for bulk add and bulk delete. Signed-off-by: Maciej Gajdzica Acked-by: Cristian Dumitrescu --- app/test/test_table_acl.c | 166 ++ 1 file changed, 166 insertions(+) diff --git a/app/test/test_t

[dpdk-dev] [PATCH v4 2/5] pipeline: added bulk add/delete functions for table

From: Marcin Kerlin Added functions for adding/deleting multiple records to table owned by pipeline. The LIBABIVER number is incremented for table and pipeline libraries. Signed-off-by: Maciej Gajdzica Signed-off-by: Marcin Kerlin Acked-by: Cristian Dumitrescu --- lib/librte_pipeline/Makefil

[dpdk-dev] [PATCH v4 1/5] table: added bulk add/delete functions for table

From: Marcin Kerlin New functions prototypes for bulk add/delete added to table API. New functions allows adding/deleting multiple records with single function call. For now those functions are implemented only for ACL table. For other tables these function pointers are set to NULL. Signed-off-b

[dpdk-dev] [PATCH v4 0/5] pipeline: add bulk add/delete functions for table

From: Marcin Kerlin This patch adds bulk add/delete functions for tables used by pipelines. It allows for adding/deleting many rules to pipeline tables in one function call. It is particulary useful for firewall pipeline which is using ACL table. After every add or delete, table is rebuild which

[dpdk-dev] [PATCH 0/4]librte_port: modify macros to access packet meta-data

> -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon > Sent: Monday, October 19, 2015 6:00 PM > To: Zhang, Roy Fan > Cc: dev at dpdk.org > Subject: Re: [dpdk-dev] [PATCH 0/4]librte_port: modify macros to access > packet meta-data > > 2015-09-11

[dpdk-dev] [PATCH v6 03/13] vhost: vring queue setup for multiple queue support

On Tue, Oct 20, 2015 at 08:49:38AM +0200, Thomas Monjalon wrote: > 2015-10-09 13:46, Yuanhan Liu: > > All queue pairs, including the default (the first) queue pair, > > are allocated dynamically, when a vring_call message is received > > first time for a specific queue pair. > > > > This is a refa

[dpdk-dev] Unit for tx_rate in rte_eth_set_queue_rate_limit?

2015-10-20 14:20, Simon K?gstr?m: > Hi! > > What is the unit of the tx_rate parameter to the > rte_eth_set_queue_rate_limit function? It's documented as > > /** > * Set the rate limitation for a queue on an Ethernet device. > * > * @param port_id > * The port identifier of the Ethernet devi

[dpdk-dev] [PATCH v2 6/8] example/ip_pipeline: add parse_hex_string for internal use

> -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon > Sent: Monday, October 19, 2015 4:50 PM > To: Singh, Jasvinder > Cc: dev at dpdk.org > Subject: Re: [dpdk-dev] [PATCH v2 6/8] example/ip_pipeline: add > parse_hex_string for internal use > >

[dpdk-dev] [PATCH v6 09/13] vhost: add API bind a virtq to a specific core

On Tue, Oct 20, 2015 at 06:16:23AM +, Xie, Huawei wrote: > On 10/9/2015 1:46 PM, Yuanhan Liu wrote: > > From: Changchun Ouyang > > > > The new API rte_vhost_core_id_set() is to bind a virtq to a specific > > core, while the another API rte_vhost_core_id_get() is for getting > > the bind core f

[dpdk-dev] Unit for tx_rate in rte_eth_set_queue_rate_limit?

Hi! What is the unit of the tx_rate parameter to the rte_eth_set_queue_rate_limit function? It's documented as /** * Set the rate limitation for a queue on an Ethernet device. * * @param port_id * The port identifier of the Ethernet device. * @param queue_idx * The queue id. * @param t

[dpdk-dev] [PATCH v2 2/2] app/testpmd: add test commands for selecting different GRE key sizes

Test commands are added to support selecting differnt length of GRE key. Signed-off-by: Helin Zhang Signed-off-by: Andrey Chilikin --- app/test-pmd/cmdline.c | 52 ++ 1 file changed, 52 insertions(+) diff --git a/app/test-pmd/cmdline.c b/app/test

[dpdk-dev] [PATCH v2 1/2] i40e: add selecting GRE key length

By default, only first 3 bytes of GRE key will be used for hash or FD calculation. With these changes, it can select 3 or 4 bytes of GRE key for hash or FD calculation. Signed-off-by: Helin Zhang Signed-off-by: Andrey Chilikin --- drivers/net/i40e/i40e_ethdev.c | 87 +++

[dpdk-dev] [PATCH v2 0/2] add selecting different GRE key length

By default, only 3 bytes of GRE key will be used for hash or FD calculation. Here adds selecting 3 or 4 bytes of GRE key for those purpose. Note that these patches are based on another patch set of "[PATCH v4 0/2] i40e: RSS/FD granularity configuration"! v2 changes: Updated with RSS/FD granularity

[dpdk-dev] [RFC PATCH v2] vhost: Add VHOST PMD

> > On 2015/09/24 2:47, Loftus, Ciara wrote: > >> The patch introduces a new PMD. This PMD is implemented as thin > wrapper > >> of librte_vhost. It means librte_vhost is also needed to compile the PMD. > >> The PMD can have 'iface' parameter like below to specify a path to > connect > >> to a vir

[dpdk-dev] [PATCH v2 6/6] ixgbe: implementation for fdir new modes' config

Hi Wenzhuo, Few questions/comments from me, see below. Thanks Konstantin > -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Wenzhuo Lu > Sent: Tuesday, September 29, 2015 6:31 AM > To: dev at dpdk.org > Subject: [dpdk-dev] [PATCH v2 6/6] ixgbe: implementation fo

[dpdk-dev] Issues with Pktgen working

> > > ./app/pktgen -c 0x1f -n 3 -- -P -m "[1:3].0, [2:4].1 > > Looks like the closing quote is missing: ./app/pktgen -c 0x1f -n 3 -- -P -m "[1:3].0,[2:4].1" - jkt

[dpdk-dev] [PATCH v6 00/13] vhost-user multiple queues enabling

On 10/09/2015 08:45 AM, Yuanhan Liu wrote: > This patch set enables vhost-user multiple queues. > > Overview > > > It depends on some QEMU patches that has already been merged to upstream. > Those qemu patches introduce some new vhost-user messages, for vhost-user > mq enabling negotiation

[dpdk-dev] [PATCH] lib: rte_*_create gives NULL/EEXIST on duped name

Also fixed a bug in many of them where if the rte_malloc of the TAILQ fails, then we return a pointer to some arbitrary existing struct. Signed-off-by: Yoni Fogel --- lib/librte_acl/rte_acl.c | 53 +-- lib/librte_hash/rte_cuckoo_hash.c | 6 +++-- lib

[dpdk-dev] [PATCH] librte_reorder: fix failing unit test cases

Earlier librte_reorder changes for support of "random sequence number as starting point" caused unit tests failure. This pacth fixes unit tests failures. Increased reorder buffer size in examples/packet_ordering/mainc.c to avoaid random traffic drops that are seen when single core and 4 ports are

[dpdk-dev] Issues with Pktgen working

Make sure the cut/paste command is not getting broken into two lines. The > normal means the shell is expecting you to finish the line after the quote. Sent from my iPhone On Oct 19, 2015, at 10:04 PM, shifalijshetty . mailto:shifalijshetty at gmail.com>> wrote: Hi , I have installed pktgen in

[dpdk-dev] [PATCH] doc: remove dejavu font from doc build

Remove requirement for occasionally hard to find/install DejaVuSansMono font. It isn't gnerally required anyway. The default mono font is sufficient. Signed-off-by: John McNamara --- doc/guides/conf.py | 1 - 1 file changed, 1 deletion(-) diff --git a/doc/guides/conf.py b/doc/guides/conf.py ind

[dpdk-dev] [PATCH] doc: change sphinx theme to the read the docs theme

Change the Sphinx default theme from "alabaster" to the ReadTheDocs theme. See for example: http://dpdk.readthedocs.org/en/latest/ This looks better for technical documentation and in particular it has a 80 char wide verbatim block rendering. Also turn off option for distracting Html mouseov

[dpdk-dev] [PATCH v2] doc: add contributors guide

Add a document to explain the DPDK patch submission and review process. Signed-off-by: John McNamara --- v2: * Fixes for mailing list comments. * Fix for broken link target. doc/guides/contributing/documentation.rst | 2 +- doc/guides/contributing/index.rst | 1 + doc/guides/contri

[dpdk-dev] [PATCH v3 6/7] virtio: simple tx routine

On Tue, 20 Oct 2015 23:30:06 +0800 Huawei Xie wrote: > + desc_idx = (uint16_t)(vq->vq_used_cons_idx & > + ((vq->vq_nentries >> 1) - 1)); > + free[0] = (struct rte_mbuf *)vq->vq_descx[desc_idx++].cookie; > + nb_free = 1; > + > + for (i = 1; i < VIRTIO_TX_FREE_NR; i++) {

[dpdk-dev] [PATCH 1/2] virtio: fixed segmentation fault in queue_release

On Tue, 20 Oct 2015 16:37:41 +0100 Bernard Iremonger wrote: > if input parameter vq is NULL, hw = vq->hw, causes a segmentation fault. > > Signed-off-by: Bernard Iremonger > --- > drivers/net/virtio/virtio_ethdev.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/dri

[dpdk-dev] [PATCH v4 2/2] app/testpmd: add test commands for RSS/FD granularity

Test commands are added to support selecting input set, or adding new input set per different pctype. Signed-off-by: Helin Zhang Signed-off-by: Andrey Chilikin --- app/test-pmd/cmdline.c | 196 + 1 file changed, 196 insertions(+) v3 changes: Supp

[dpdk-dev] [PATCH v4 1/2] i40e: RSS/FD granularity configuration

The default fields of a received packet are loaded from firmware, which cannot be modified even users want to use different fields for RSS or flow director. Here adds more flexibilities of selecting packet fields for hash calculation or flow director for users. Signed-off-by: Helin Zhang Signed-o

[dpdk-dev] [PATCH v4 0/2] i40e: RSS/FD granularity configuration

The default fields of a received packet are loaded from firmware, which cannot be modified even users want to use different fields for RSS or flow director. Here adds more flexibilities of selecting or adding packet fields for hash calculation or flow director to users. It also includes the modific

[dpdk-dev] [PATCH] doc: Add missing new line before code block

The patch adds missing new line to "Managing ABI updates" section. Signed-off-by: Tetsuya Mukawa --- doc/guides/contributing/versioning.rst | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/doc/guides/contributing/versioning.rst b/doc/guides/contributing/versioning.rst index

[dpdk-dev] [PATCH v2 1/1] ethdev: remove the imissed deprecation tag

Remove the deprecation tag and notice for imissed as it is a generic register that accounts for packets that were dropped by the HW, because there are no available mbufs (RX queues are full). imissed is different to ierrors and can help with general debug. Fixes: 49f386542af4 ("ethdev: remove driv

[dpdk-dev] [PATCH v3 02/20] librte_ether: add fields from rte_pci_driver to rte_eth_dev_data

Hi Michael, > > @@ -1795,8 +1790,7 @@ rte_eth_dev_info_get(uint8_t port_id, struct > rte_eth_dev_info *dev_info) > > FUNC_PTR_OR_RET(*dev->dev_ops->dev_infos_get); > > (*dev->dev_ops->dev_infos_get)(dev, dev_info); > > dev_info->pci_dev = dev->pci_dev; > > Here also pci_dev, I thin

[dpdk-dev] [PATCH] ixgbe: prefetch packet headers in vector PMD receive function

On 19/10/15 19:57, Ananyev, Konstantin wrote: > > >> -Original Message- >> From: Zoltan Kiss [mailto:zoltan.kiss at linaro.org] >> Sent: Monday, October 19, 2015 5:30 PM >> To: Ananyev, Konstantin; Richardson, Bruce; dev at dpdk.org >> Subject: Re: [PATCH] ixgbe: prefetch packet headers i

[dpdk-dev] [PATCH] doc: add contributors guide

> -Original Message- > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com] > Sent: Thursday, October 15, 2015 10:37 PM > To: Mcnamara, John > Cc: dev at dpdk.org > Subject: Re: [dpdk-dev] [PATCH] doc: add contributors guide > > Hi John, > > 2015-10-15 17:51, John McNamara: > > Add

[dpdk-dev] [PATCH v3 02/20] librte_ether: add fields from rte_pci_driver to rte_eth_dev_data

Hi Michael, > -Original Message- > From: Qiu, Michael > Sent: Tuesday, October 20, 2015 10:19 AM > To: Iremonger, Bernard; dev at dpdk.org > Subject: Re: [dpdk-dev] [PATCH v3 02/20] librte_ether: add fields from > rte_pci_driver to rte_eth_dev_data > > On 2015/10/13 0:26, Bernard Iremonge

[dpdk-dev] [PATCH v3] ixgbe: remove rx jabber from ierrors

Remove receive jabber count (rjc) from ierrors count as the register overlaps with the CRC error register, previously causing some packets to be counted twice. Signed-off-by: Harry van Haaren --- v3: Add details about register overlap v2: Fix typo drivers/net/ixgbe/ixgbe_ethdev.c | 1 - 1 file

[dpdk-dev] dpdk proposal installation process

On Tue, Oct 20, 2015 at 12:21:00AM +, Arevalo, Mario Alfredo C wrote: > Hi folks, > > Good day, this is a proposal in order to improve the dpdk install > process, > I would like to know your point of view about the next points according to > previous conversations :) in order to create

[dpdk-dev] [dpdk-dev, PATCHv5, 1/8] ethdev: add new API to retrieve RX/TX queue information

On 20/10/2015 09:53, Qiu, Michael wrote: > But as I know it is different all the time, am I right? > If yes, I don't know what's the value of this field. It can be used to get some snapshot/instant view informations while we have to monitor and debug.

[dpdk-dev] [PATCH v5 3/4] vhost: using EVENTFD_COPY2

Thanks Pavel for this work. This is what we think is the better implementation for eventfd proxy, in our last review. Could you add an additional patch to remove the old implementation? Again, please run checkpatch.pl against your patch. On 8/29/2015 2:51 AM, Pavel Boldin wrote: [...] > + > +int

[dpdk-dev] [dpdk-dev, PATCHv6 1/6] ethdev: enhance rte_eth_(tx|rx)q_info struct

> -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen Hemminger > Sent: Monday, October 19, 2015 11:44 PM > To: Amine Kherbouche > Cc: dev at dpdk.org > Subject: Re: [dpdk-dev] [dpdk-dev, PATCHv6 1/6] ethdev: enhance > rte_eth_(tx|rx)q_info struct > > On

[dpdk-dev] When I run test-pmd, most of received packets(loop-backed packet) have RX-error.

On Tue, Oct 20, 2015 at 2:12 AM, ??? wrote: > Dear DPDK experts. > > Thank you very much for your best great efforts and precious answers. > > > When I run test-pmd, most of received packets are RX-error. > > The computer has two 10GbE ports Intel NIC and the two ports are loop-backed > each othe

[dpdk-dev] [PATCH v6 10/13] ixgbe: support VMDq RSS in non-SRIOV environment

Helin, Konstantin, could you review please? 2015-10-09 13:46, Yuanhan Liu: > From: Changchun Ouyang > > In non-SRIOV environment, VMDq RSS could be enabled by MRQC register. > In theory, the queue number per pool could be 2 or 4, but only 2 queues > are available due to HW limitation, the same l

[dpdk-dev] [PATCH v5 2/4] vhost: add EVENTFD_COPY2 ioctl

On 8/29/2015 2:51 AM, Pavel Boldin wrote: > Signed-off-by: Pavel Boldin > --- > OK with the patch. Please run checkpatch.pl against your patch.

[dpdk-dev] [PATCH v3] ixgbe: remove rx jabber from ierrors

> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Harry van Haaren > Sent: Tuesday, October 20, 2015 10:23 AM > To: dev at dpdk.org > Subject: [dpdk-dev] [PATCH v3] ixgbe: remove rx jabber from ierrors > > Remove receive jabber count (rjc) from ierrors count as the register overlaps > with

[dpdk-dev] [dpdk-dev, PATCHv6 1/6] ethdev: enhance rte_eth_(tx|rx)q_info struct

Hi Amine, > -Original Message- > From: Amine Kherbouche [mailto:amine.kherbouche at 6wind.com] > Sent: Monday, October 19, 2015 11:06 PM > To: dev at dpdk.org > Cc: amine.kherbouche at 6wind.com; vincent.jardin at 6wind.com; Ananyev, > Konstantin > Subject: [dpdk-dev,PATCHv6 1/6] ethdev:

[dpdk-dev] [PATCH v6 06/13] virtio: read virtio_net_config correctly

2015-10-20 15:23, Yuanhan Liu: > On Tue, Oct 20, 2015 at 09:07:49AM +0200, Thomas Monjalon wrote: > > 2015-10-09 13:46, Yuanhan Liu: > > > From: Changchun Ouyang > > > > > > The old code adjusts the config bytes we want to read depending on > > > what kind of features we have, but we later cast t

[dpdk-dev] [PATCH v6 04/13] vhost: rxtx: prepare work for multiple queue support

2015-10-20 15:10, Yuanhan Liu: > On Tue, Oct 20, 2015 at 08:57:52AM +0200, Thomas Monjalon wrote: > > > - if (unlikely(queue_id != VIRTIO_RXQ)) { > > > - LOG_DEBUG(VHOST_DATA, "mq isn't supported in this version.\n"); > > > + if (unlikely(!is_valid_virt_queue_idx(queue_id, 0, dev->virt_qp_n

[dpdk-dev] [PATCH v3 02/20] librte_ether: add fields from rte_pci_driver to rte_eth_dev_data

On 2015/10/13 0:26, Bernard Iremonger wrote: > add dev_flags to rte_eth_dev_data, add macros for dev_flags. > add kdrv to rte_eth_dev_data. > add numa_node to rte_eth_dev_data. > add drv_name to rte_eth_dev_data. > use dev_type to distinguish between vdev's and pdev's. > remove pci_dev branches. >

[dpdk-dev] [PATCH 3/5] bnx2x: Add RTE_LIBRTE_BNX2X_REG_ACCESS config option

2015-09-11 17:42, Rasesh Mody: > From: Harish Patil > --- a/config/common_linuxapp > +++ b/config/common_linuxapp > @@ -220,6 +220,7 @@ CONFIG_RTE_LIBRTE_BNX2X_DEBUG_INIT=n > CONFIG_RTE_LIBRTE_BNX2X_DEBUG_RX=n > CONFIG_RTE_LIBRTE_BNX2X_DEBUG_TX=n > CONFIG_RTE_LIBRTE_BNX2X_MF_SUPPORT=n > +CONFIG

[dpdk-dev] [PATCH v6 09/13] vhost: add API bind a virtq to a specific core

2015-10-20 14:25, Yuanhan Liu: > On Tue, Oct 20, 2015 at 06:16:23AM +, Xie, Huawei wrote: > > On 10/9/2015 1:46 PM, Yuanhan Liu wrote: > > > From: Changchun Ouyang > > > > > > The new API rte_vhost_core_id_set() is to bind a virtq to a specific > > > core, while the another API rte_vhost_core_

[dpdk-dev] [PATCH v6 06/13] virtio: read virtio_net_config correctly

2015-10-09 13:46, Yuanhan Liu: > From: Changchun Ouyang > > The old code adjusts the config bytes we want to read depending on > what kind of features we have, but we later cast the entire buf we > read with "struct virtio_net_config", which is obviously wrong. When describing a bug, it is impor

[dpdk-dev] [PATCH v5 1/4] vhost: eventfd_link: refactoring EVENTFD_COPY handler

On 8/29/2015 2:51 AM, Pavel Boldin wrote: > * Move ioctl `EVENTFD_COPY' code to a separate function > * Remove extra #includes > * Introduce function fget_from_files > * Fix ioctl return values > > Signed-off-by: Pavel Boldin Basically OK with this patch since we have reviewed before. Could you sp

[dpdk-dev] [PATCH v6 05/13] vhost-user: handle VHOST_USER_RESET_OWNER correctly

2015-10-09 13:46, Yuanhan Liu: > Destroy corresponding device when a VHOST_USER_RESET_OWNER message is > received, otherwise, the vhost-switch would still try to access vq > of that device, which results to SIGSEG fault, and let vhost-switch > crash in the end. It is a fix, so the title should loo

  1   2   >