Re: [dpdk-dev] [PATCH 5/6] net/mlx5: implement descriptor status API

Hi Olivier, On Wed, Mar 01, 2017 at 06:19:11PM +0100, Olivier Matz wrote: > Since there is no "descriptor done" flag like on Intel drivers, the > approach is different on mlx5 driver. > - for Tx, we call txq_complete() to free descriptors processed by > the hw, then we check if the descriptor is

[dpdk-dev] [PATCH v2 3/3] lib/librte_ether: fix code style issues

Fixed code style issues following Yuanhan Liu's review. Signed-off-by: Ami Sabo --- lib/librte_ether/rte_ethdev.h | 2 +- lib/librte_ether/rte_ether_version.map | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_

[dpdk-dev] [PATCH v2 0/3] Fix virtio-user multi-process crash.

The patchset fixes secondary process crash issue when it tries to access virtio-user pmd (e.g. via rte_eth_rx_burst). The crash happens because the secondary process calls, at virtio_user_pmd_probe() to virtio_user_eth_dev_alloc()-> rte_eth_dev_allocate() instead of eth_dev_attach_secondary(), as

[dpdk-dev] [PATCH v2 1/3] lib/librte_ether: export secondary attach function

Today eth_dev_attach_secondary is defined as static and can only be called by pci drivers. However, the functionality is also required for non-pci drivers - so the patch export the function. Signed-off-by: Ami Sabo --- lib/librte_ether/rte_ethdev.c | 6 +++--- lib/librte_ether/rte_ethd

[dpdk-dev] [PATCH v2 2/3] net/virtio-user: fix multi-process issue

Secondary process doesn't properly attach to the rte_eth_device initialized by the primary process. ccessing device from secondary process (e.g. via rte_eth_rx_burst), causes process to crash. because rte_eth_dev_data is not properly set. The issue was flood by 'commit 7f95f78a8aea ("ethdev: clea

[dpdk-dev] [PATCH 13/13] net/sfc: support Rx packed stream EF10-specific datapath

Signed-off-by: Andrew Rybchenko --- doc/guides/nics/sfc_efx.rst | 23 +- drivers/net/sfc/Makefile | 1 + drivers/net/sfc/efsys.h | 2 +- drivers/net/sfc/sfc_dp.h | 5 +- drivers/net/sfc/sfc_dp_rx.h | 8 + drivers/net/sfc/sfc_ef10_ps_rx.c | 659 +

[dpdk-dev] [PATCH 06/13] net/sfc: implement EF10 native Rx datapath

Signed-off-by: Andrew Rybchenko --- doc/guides/nics/sfc_efx.rst | 5 +- drivers/net/sfc/Makefile | 1 + drivers/net/sfc/sfc_dp.h | 1 + drivers/net/sfc/sfc_dp_rx.h | 22 ++ drivers/net/sfc/sfc_ef10_rx.c | 713 ++ drivers/net/sfc/sfc_eth

[dpdk-dev] [PATCH 04/13] net/sfc: factor out libefx-based Rx datapath

Split control and datapath to make datapath substitutable and possibly reusable with alternative control path. libefx-based Rx datapath is bound to libefx control path, but other datapaths should be possible to use with alternative control path(s). Signed-off-by: Andrew Rybchenko --- doc/guides

[dpdk-dev] [PATCH 07/13] net/sfc: factory out libefx-based Tx datapath

Split control and datapath to make datapath substitutable and possibly reusable with alternative control path. libefx-based Tx datapath is bound to libefx control path, but other datapaths should be possible to use with alternative control path(s). Signed-off-by: Andrew Rybchenko --- doc/guides

[dpdk-dev] [PATCH 10/13] net/sfc: implement EF10 native Tx datapath

Signed-off-by: Andrew Rybchenko --- doc/guides/nics/sfc_efx.rst | 5 +- drivers/net/sfc/Makefile | 1 + drivers/net/sfc/sfc_dp_tx.h | 17 ++ drivers/net/sfc/sfc_ef10_tx.c | 439 ++ drivers/net/sfc/sfc_ethdev.c | 1 + drivers/net/sfc/sfc_ev.

[dpdk-dev] [PATCH 12/13] net/sfc: implement simple EF10 native Tx datapath

The datapath does not support VLAN insertion, TSO and multi-segment mbufs. Signed-off-by: Andrew Rybchenko --- doc/guides/nics/sfc_efx.rst | 5 ++- drivers/net/sfc/sfc_dp_tx.h | 1 + drivers/net/sfc/sfc_ef10_tx.c | 78 +++ drivers/net/sfc/sfc_ethdev.

[dpdk-dev] [PATCH 05/13] net/sfc: Rx scatter is a datapath-dependent feature

Signed-off-by: Andrew Rybchenko --- drivers/net/sfc/sfc_dp_rx.h | 2 ++ drivers/net/sfc/sfc_rx.c| 8 2 files changed, 10 insertions(+) diff --git a/drivers/net/sfc/sfc_dp_rx.h b/drivers/net/sfc/sfc_dp_rx.h index 5a714a1..66d655f 100644 --- a/drivers/net/sfc/sfc_dp_rx.h +++ b/drivers

[dpdk-dev] [PATCH 09/13] net/sfc: TSO is a datapath dependent feature

Signed-off-by: Andrew Rybchenko --- drivers/net/sfc/sfc_dp_tx.h | 1 + drivers/net/sfc/sfc_tx.c| 6 +- 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/sfc/sfc_dp_tx.h b/drivers/net/sfc/sfc_dp_tx.h index 9f9948f..b6b7084 100644 --- a/drivers/net/sfc/sfc_dp_tx.h ++

[dpdk-dev] [PATCH 11/13] net/sfc: multi-segment support as is Tx datapath features

Signed-off-by: Andrew Rybchenko --- drivers/net/sfc/sfc_dp_tx.h | 1 + drivers/net/sfc/sfc_ef10_tx.c | 2 +- drivers/net/sfc/sfc_ethdev.c | 3 +++ drivers/net/sfc/sfc_tx.c | 10 +- 4 files changed, 14 insertions(+), 2 deletions(-) diff --git a/drivers/net/sfc/sfc_dp_tx.h b/dri

[dpdk-dev] [PATCH 08/13] net/sfc: VLAN insertion is a datapath dependent feature

Signed-off-by: Andrew Rybchenko --- drivers/net/sfc/sfc_dp_tx.h | 2 ++ drivers/net/sfc/sfc_ethdev.c | 3 ++- drivers/net/sfc/sfc_tx.c | 14 +++--- 3 files changed, 15 insertions(+), 4 deletions(-) diff --git a/drivers/net/sfc/sfc_dp_tx.h b/drivers/net/sfc/sfc_dp_tx.h index 4879db

[dpdk-dev] [PATCH 03/13] net/sfc: do not use Rx queue control state on datapath

Rx queue flags should keep the information required on datapath. It is a preparation to split control and data paths. Signed-off-by: Andrew Rybchenko --- drivers/net/sfc/sfc_ev.c | 5 +++-- drivers/net/sfc/sfc_rx.c | 12 +++- drivers/net/sfc/sfc_rx.h | 12 +--- 3 files changed,

[dpdk-dev] [PATCH 01/13] net/sfc: callbacks should depend on EvQ usage

Use different sets of libefx EvQ callbacks for management, transmit and receive event queue. It makes event handling more robust against unexpected events. Also it is required for alternative datapath support. Signed-off-by: Andrew Rybchenko --- drivers/net/sfc/sfc_ev.c | 107 ++

[dpdk-dev] [PATCH 02/13] net/sfc: emphasis that RSS hash flag is an Rx queue flag

Style fix to establish namespace for Rx queue flag defines. Signed-off-by: Andrew Rybchenko --- drivers/net/sfc/sfc_rx.c | 4 ++-- drivers/net/sfc/sfc_rx.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/sfc/sfc_rx.c b/drivers/net/sfc/sfc_rx.c index 906536e..e7

[dpdk-dev] [PATCH 00/13] Improve Solarflare PMD performance

Implement EF10 (SFN7xxx and SFN8xxx) native datapaths which may be chosen per device using PCI whitelist device arguments. libefx-based datapath implementation is bound to API and structure imposed by the libefx. It has many indirect function calls to provide HW abstraction (bad for CPU pipeline)

[dpdk-dev] [PATCH] vhost: remove a hack on queue allocation

We used to allocate queues based on the index from SET_VRING_CALL request: if corresponding queue hasn't been allocated, allocate it. Though it's pratically right (it's the first per-vring request we will get from QEMU for vhost-user negotiation), but it's not technically right: it's not documente

[dpdk-dev] [PATCH] vhost: remove a hack on queue allocation

We used to allocate queues based on the index from SET_VRING_CALL request: if corresponding queue hasn't been allocated, allocate it. Though it's pratically right (it's the first per-vring request we will get from QEMU for vhost-user negotiation), but it's not technically right: it's not documente

[dpdk-dev] [PATCH] eventdev: Fix links_map initialization

This patch initializes the links_map array entries to EVENT_QUEUE_SERVICE_PRIORITY_INVALID, as expected by rte_event_port_links_get(). Signed-off-by: Gage Eads --- lib/librte_eventdev/rte_eventdev.c | 17 - 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/lib/librte

Re: [dpdk-dev] [PATCH 3/6] net/e1000: implement descriptor status API (igb)

Hi Olivier, > -Original Message- > From: Olivier Matz [mailto:olivier.m...@6wind.com] > Sent: Thursday, March 2, 2017 1:19 AM > To: dev@dpdk.org; thomas.monja...@6wind.com; Ananyev, Konstantin; Lu, > Wenzhuo; Zhang, Helin; Wu, Jingjing; adrien.mazarg...@6wind.com; > nelio.laranje...@6wind.

Re: [dpdk-dev] [PATCH 4/6] net/e1000: implement descriptor status API (em)

Hi Oliver, > -Original Message- > From: Olivier Matz [mailto:olivier.m...@6wind.com] > Sent: Thursday, March 2, 2017 1:19 AM > To: dev@dpdk.org; thomas.monja...@6wind.com; Ananyev, Konstantin; Lu, > Wenzhuo; Zhang, Helin; Wu, Jingjing; adrien.mazarg...@6wind.com; > nelio.laranje...@6wind.c

Re: [dpdk-dev] [PATCH v3 15/16] net/avp: device start and stop operations

On Wed, 1 Mar 2017 19:20:07 -0500 Allain Legacy wrote: > + > +static void > +avp_dev_close(struct rte_eth_dev *eth_dev) > +{ > + struct avp_dev *avp = AVP_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private); > + int ret; > + > + rte_spinlock_lock(&avp->lock); > + if (avp->flags & AVP_F_

Re: [dpdk-dev] [PATCH v3 13/16] net/avp: device statistics operations

On Wed, 1 Mar 2017 19:20:05 -0500 Allain Legacy wrote: > +static void > +avp_dev_stats_get(struct rte_eth_dev *eth_dev, struct rte_eth_stats *stats) > +{ > + struct avp_dev *avp = AVP_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private); > + unsigned int i; > + > + memset(stats, 0, sizeof(*s

[dpdk-dev] [PATCH v3 16/16] doc: adds information related to the AVP PMD

Updates the documentation and feature lists for the AVP PMD device. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters Acked-by: John McNamara --- MAINTAINERS| 1 + doc/guides/nics/avp.rst| 112 + doc/guides/nic

[dpdk-dev] [PATCH v3 11/16] net/avp: packet receive functions

Adds function required for receiving packets from the host application via AVP device queues. Both the simple and scattered functions are supported. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- drivers/net/avp/Makefile | 1 + drivers/net/avp/avp_ethdev.c | 461

[dpdk-dev] [PATCH v3 15/16] net/avp: device start and stop operations

Adds support for device start and stop functions. This allows an application to control the administrative state of an AVP device. Stopping the device will notify the host application to stop sending packets on that device's receive queues. Signed-off-by: Allain Legacy Signed-off-by: Matt Peter

[dpdk-dev] [PATCH v3 13/16] net/avp: device statistics operations

Adds device functions to query and reset statistics. Signed-off-by: Allain Legacy --- drivers/net/avp/avp_ethdev.c | 68 1 file changed, 68 insertions(+) diff --git a/drivers/net/avp/avp_ethdev.c b/drivers/net/avp/avp_ethdev.c index bd7f7ec..81c6551

[dpdk-dev] [PATCH v3 12/16] net/avp: packet transmit functions

Adds support for packet transmit functions so that an application can send packets to the host application via an AVP device queue. Both the simple and scattered functions are supported. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- drivers/net/avp/avp_ethdev.c | 349

[dpdk-dev] [PATCH v3 14/16] net/avp: device promiscuous functions

Adds support for setting and clearing promiscuous mode on an AVP device. When enabled the _mac_filter function will allow packets destined to any MAC address to be processed by the receive functions. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- drivers/net/avp/avp_ethdev.c | 32 +

[dpdk-dev] [PATCH v3 10/16] net/avp: queue setup and release

Adds queue management operations so that an appliation can setup and release the transmit and receive queues. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- drivers/net/avp/avp_ethdev.c | 144 ++- 1 file changed, 143 insertions(+), 1 deletion

[dpdk-dev] [PATCH v3 09/16] net/avp: device configuration

Adds support for "dev_configure" operations to allow an application to configure the device. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- drivers/net/avp/avp_ethdev.c | 131 +++ 1 file changed, 131 insertions(+) diff --git a/drivers/net/av

[dpdk-dev] [PATCH v3 04/16] net/avp: add PMD version map file

Adds a default ABI version file for the AVP PMD. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- drivers/net/avp/rte_pmd_avp_version.map | 4 1 file changed, 4 insertions(+) create mode 100644 drivers/net/avp/rte_pmd_avp_version.map diff --git a/drivers/net/avp/rte_pmd_avp_ve

[dpdk-dev] [PATCH v3 07/16] net/avp: driver registration

Adds the initial framework for registering the driver against the support PCI device identifiers. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- config/common_linuxapp | 1 + config/defconfig_i686-native-linuxapp-gcc| 5 + config/defconfig_i686-native-

[dpdk-dev] [PATCH v3 05/16] net/avp: debug log macros

Adds a header file with log macros for the AVP PMD Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- config/common_base | 4 drivers/net/avp/avp_logs.h | 59 ++ 2 files changed, 63 insertions(+) create mode 100644 drivers/net/

[dpdk-dev] [PATCH v3 08/16] net/avp: device initialization

Adds support for initialization newly probed AVP PCI devices. Initial queue translations are setup in preparation for device configuration. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- drivers/net/avp/avp_ethdev.c | 733 +++ 1 file changed

[dpdk-dev] [PATCH v3 02/16] net/avp: public header files

Adds public/exported header files for the AVP PMD. The AVP device is a shared memory based device. The structures and constants that define the method of operation of the device must be visible by both the PMD and the host DPDK application. They must not change without proper version controls an

[dpdk-dev] [PATCH v3 06/16] drivers/net: adds driver makefiles for AVP PMD

Adds a default Makefile to the driver directory but does not include any source files. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- drivers/net/Makefile | 1 + drivers/net/avp/Makefile | 52 2 files changed, 53 insertions(+)

[dpdk-dev] [PATCH v3 01/16] config: adds attributes for the AVP PMD

Updates the common base configuration file to include a top level config attribute for the AVP PMD. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- config/common_base | 5 + 1 file changed, 5 insertions(+) diff --git a/config/common_base b/config/common_base index aeee13e..912b

[dpdk-dev] [PATCH v3 03/16] maintainers: claim responsibility for AVP PMD

Updating the maintainers file to claim the AVP PMD driver on behalf of Wind River Systems, Inc. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters --- MAINTAINERS | 5 + 1 file changed, 5 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 5030c1c..fef23a0 100644 --- a/MAINTAIN

[dpdk-dev] [PATCH v3 00/16] Wind River Systems AVP PMD

This patch series submits an initial version of the AVP PMD from Wind River Systems. The series includes shared header files, driver implementation, and changes to documentation files in support of this new driver. The AVP driver is a shared memory based device. It is intended to be used as a PM

Re: [dpdk-dev] [PATCH 1/2] crypto/qat: add ZUC EEA3 cipher capability

Hi Arek, > -Original Message- > From: Kusztal, ArkadiuszX > Sent: Wednesday, March 01, 2017 7:56 AM > To: dev@dpdk.org > Cc: Trahe, Fiona; De Lara Guarch, Pablo; Griffin, John; Jain, Deepak K; > Kusztal, ArkadiuszX > Subject: [PATCH 1/2] crypto/qat: add ZUC EEA3 cipher capability > > This

Re: [dpdk-dev] [PATCH v5] doc: use corelist instead of coremask

> On Mar 1, 2017, at 1:40 PM, Thomas Monjalon wrote: > > 2017-02-28 17:14, Mcnamara, John: >> >>> The coremask option in DPDK is difficult to use and we should be >>> promoting the use of the corelist (-l) option. The patch >>> adjusts the docs to use -l EAL option instead of the -c option. >>>

Re: [dpdk-dev] [PATCH v5] doc: use corelist instead of coremask

2017-02-28 17:14, Mcnamara, John: > > > The coremask option in DPDK is difficult to use and we should be > > promoting the use of the corelist (-l) option. The patch > > adjusts the docs to use -l EAL option instead of the -c option. > > > > The patch only changes the docs and not the code as the

Re: [dpdk-dev] [PATCH v2 1/2] librte_net: add crc init and compute APIs

2017-02-28 12:08, Jasvinder Singh: > lib/librte_net/rte_net_crc.c | 664 > + > lib/librte_net/rte_net_crc.h | 101 ++ I think it should be in librte_hash. Please check lib/librte_hash/rte_hash_crc.h

Re: [dpdk-dev] [dpdk-stable] [PATCH] net/i40e: Fix a typo in i40e_flow.c.

On 2/28/2017 2:32 AM, Yuanhan Liu wrote: > On Mon, Feb 27, 2017 at 04:31:28PM +, Ferruh Yigit wrote: >> On 2/27/2017 3:50 PM, Mcnamara, John wrote: >>> >>> -Original Message- From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Rami Rosen Sent: Monday, February 27, 2017 4

Re: [dpdk-dev] [PATCH] devtools: add long lines to the list of options

2017-03-01 10:08, Stephen Hemminger: > > --- a/devtools/checkpatches.sh > > +++ b/devtools/checkpatches.sh > > @@ -45,7 +45,7 @@ options="$options > > --ignore=LINUX_VERSION_CODE,FILE_PATH_CHANGES,\ > > VOLATILE,PREFER_PACKED,PREFER_ALIGNED,PREFER_PRINTF,\ > > PREFER_KERNEL_TYPES,BIT_MACRO,CONST

Re: [dpdk-dev] [PATCH 1/6] ethdev: add descriptor status API

On 03/01/2017 08:19 PM, Olivier Matz wrote: Introduce a new API to get the status of a descriptor. For Rx, it is almost similar to rx_descriptor_done API, except it differentiates "used" descriptors (which are hold by the driver and not returned to the hardware). For Tx, it is a new API. The d

Re: [dpdk-dev] [PATCH] devtools: add long lines to the list of options

On Wed, 1 Mar 2017 12:44:19 -0500 Allain Legacy wrote: > The checkpatch.pl tool is flagging warnings on long debug log strings. > Since splitting these strings makes it difficult to search for logs it is > preferred to allow these as exceptions to the long line rule. The addition > of the LONG_L

Re: [dpdk-dev] [PATCH 0/6] get status of Rx and Tx descriptors

On Wed, 1 Mar 2017 18:19:06 +0100 Olivier Matz wrote: > This patchset introduces a new ethdev API: > - rte_eth_rx_descriptor_status() > - rte_eth_tx_descriptor_status() > > The Rx API is aims to replace rte_eth_rx_descriptor_done() which > does almost the same, but does not differentiate the ca

Re: [dpdk-dev] [PATCH 0/6] get status of Rx and Tx descriptors

On 03/01/2017 08:19 PM, Olivier Matz wrote: This patchset introduces a new ethdev API: - rte_eth_rx_descriptor_status() - rte_eth_tx_descriptor_status() May be corresponding features should be added to the NICs documentation? The Rx API is aims to replace rte_eth_rx_descriptor_done() which do

[dpdk-dev] [PATCH] devtools: add long lines to the list of options

The checkpatch.pl tool is flagging warnings on long debug log strings. Since splitting these strings makes it difficult to search for logs it is preferred to allow these as exceptions to the long line rule. The addition of the LONG_LINE_STRINGS to the list of exceptions will allow lines that end w

Re: [dpdk-dev] [PATCH] doc: adds information related to the AVP PMD

Please ignore. Allain Legacy, Software Developer direct 613.270.2279  fax 613.492.7870 skype allain.legacy   > -Original Message- > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Allain Legacy > Sent: Wednesday, March 01, 2017 12:42 PM > To: thomas.monja...@6wind.com > Cc: dev@dpd

[dpdk-dev] [PATCH] doc: adds information related to the AVP PMD

Updates the documentation and feature lists for the AVP PMD device. Signed-off-by: Allain Legacy Signed-off-by: Matt Peters Acked-by: John McNamara --- MAINTAINERS| 1 + doc/guides/nics/avp.rst| 112 + doc/guides/nic

[dpdk-dev] [PATCH 6/6] net/i40e: implement descriptor status API

Signed-off-by: Olivier Matz --- drivers/net/i40e/i40e_ethdev.c| 2 ++ drivers/net/i40e/i40e_ethdev_vf.c | 2 ++ drivers/net/i40e/i40e_rxtx.c | 56 +++ drivers/net/i40e/i40e_rxtx.h | 4 +++ 4 files changed, 64 insertions(+) diff --git a/drivers

[dpdk-dev] [PATCH 5/6] net/mlx5: implement descriptor status API

Since there is no "descriptor done" flag like on Intel drivers, the approach is different on mlx5 driver. - for Tx, we call txq_complete() to free descriptors processed by the hw, then we check if the descriptor is between tail and head - for Rx, we need to browse the cqes, managing compressed on

[dpdk-dev] [PATCH 4/6] net/e1000: implement descriptor status API (em)

Signed-off-by: Olivier Matz --- drivers/net/e1000/e1000_ethdev.h | 5 drivers/net/e1000/em_ethdev.c| 2 ++ drivers/net/e1000/em_rxtx.c | 49 3 files changed, 56 insertions(+) diff --git a/drivers/net/e1000/e1000_ethdev.h b/drivers/net/e100

[dpdk-dev] [PATCH 3/6] net/e1000: implement descriptor status API (igb)

Signed-off-by: Olivier Matz --- drivers/net/e1000/e1000_ethdev.h | 5 + drivers/net/e1000/igb_ethdev.c | 2 ++ drivers/net/e1000/igb_rxtx.c | 46 3 files changed, 53 insertions(+) diff --git a/drivers/net/e1000/e1000_ethdev.h b/drivers/net/e10

[dpdk-dev] [PATCH 2/6] net/ixgbe: implement descriptor status API

Signed-off-by: Olivier Matz --- drivers/net/ixgbe/ixgbe_ethdev.c | 4 +++ drivers/net/ixgbe/ixgbe_ethdev.h | 5 drivers/net/ixgbe/ixgbe_rxtx.c | 55 3 files changed, 64 insertions(+) diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixg

[dpdk-dev] [PATCH 1/6] ethdev: add descriptor status API

Introduce a new API to get the status of a descriptor. For Rx, it is almost similar to rx_descriptor_done API, except it differentiates "used" descriptors (which are hold by the driver and not returned to the hardware). For Tx, it is a new API. The descriptor_done() API, and probably the rx_queu

[dpdk-dev] [PATCH 0/6] get status of Rx and Tx descriptors

This patchset introduces a new ethdev API: - rte_eth_rx_descriptor_status() - rte_eth_tx_descriptor_status() The Rx API is aims to replace rte_eth_rx_descriptor_done() which does almost the same, but does not differentiate the case of a descriptor used by the driver (not returned to the hw). The

[dpdk-dev] [PATCH v3] proc-info: added collectd-format and host-id options.

Extended proc-info application to send DPDK port statistics to STDOUT in the format expected by collectd exec plugin. Added HOST ID option to identify the host DPDK process is running on when multiple instance of DPDK are running in parallel. This is needed for the barometer project in OPNFV. Sign

[dpdk-dev] [PATCH] crypto/scheduler: add failover scheduling mode

This patch adds the failover scheduling mode to scheduler crypto PMD. In this mode the 1st and 2nd slaves are used as primary and secondary slaves respectively. The scheduler will try to enqueue the incoming crypto operations to the primary slave first, and will only enqueue to the secondary slave

Re: [dpdk-dev] [PATCH v2 11/15] net/avp: packet receive functions

> -Original Message- > From: Stephen Hemminger [mailto:step...@networkplumber.org] > > In checkpatch source there is a regex to identify logging functions and > special > exceptions for long lines etc. But the logging functions are for kernel > (printk > etc), not DPDK logging functions

Re: [dpdk-dev] [RFC 17.05 v1 0/3] Merge l3fwd-acl and l3fwd

Hi Konstantin, Thank you for the review. RSS hash value changes could be due to merge, I didn't make that change. I will go through the changes and fix it in 'v2' patch along with RFC removed and checkpatch fix. Thanks. On Tue, Feb 28, 2017 at 2:36 AM, Ananyev, Konstantin < konstantin.anan...@i

Re: [dpdk-dev] [PATCH v2 11/15] net/avp: packet receive functions

On Wed, 01 Mar 2017 15:14:57 +0100 Thomas Monjalon wrote: > 2017-03-01 13:23, Legacy, Allain: > > > -Original Message- > > > From: Bruce Richardson [mailto:bruce.richard...@intel.com] > > > In my experience, checkpatch ignores long lines that are due to error > > > messages. Perhaps you n

Re: [dpdk-dev] [PATCH v2 11/15] net/avp: packet receive functions

> -Original Message- > From: Thomas Monjalon [mailto:thomas.monja...@6wind.com] > > WARNING:LONG_LINE_STRING: line over 80 characters > > #120: FILE: drivers/net/avp/avp_ethdev.c:236: > > + PMD_DRV_LOG(ERR, "Timeout while waiting for a > response for %u\n", > > > > There

Re: [dpdk-dev] [PATCH 2/2] net/mlx5: add hardware TSO support for VXLAN and GRE

On Tue, Feb 28, 2017 at 05:53:48PM +0200, Shahaf Shuler wrote: > This commit adds support for hardware TSO for tunneled packets. > > Signed-off-by: Shahaf Shuler > --- > drivers/net/mlx5/mlx5_ethdev.c | 4 +++- > drivers/net/mlx5/mlx5_rxtx.c | 9 + > 2 files changed, 12 insertions(+),

Re: [dpdk-dev] [PATCH v8 09/18] lib: add symbol versioning to distributor

ERROR:SPACING: space prohibited before that ',' (ctx:WxW) #84: FILE: lib/librte_distributor/rte_distributor.c:172: +BIND_DEFAULT_SYMBOL(rte_distributor_get_pkt, , 17.05); ^ FYI, checkpatch does not like this regardless of whether there's a space there

[dpdk-dev] [PATCH v8 18/18] maintainers: add to distributor lib maintainers

Signed-off-by: David Hunt --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 5030c1c..42eece0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -501,6 +501,7 @@ F: doc/guides/sample_app_ug/ip_reassembly.rst Distributor M: Bruce Richardson +M: Davi

[dpdk-dev] [PATCH v8 16/18] examples/distributor: give Rx thread a core

This so that with the increased amount of stats we are counting, we don't interfere with the rx core. Signed-off-by: David Hunt --- examples/distributor/main.c | 50 ++--- 1 file changed, 34 insertions(+), 16 deletions(-) diff --git a/examples/distributor

[dpdk-dev] [PATCH v8 17/18] doc: distributor library changes for new burst API

Signed-off-by: David Hunt --- doc/guides/prog_guide/packet_distrib_lib.rst | 3 +++ 1 file changed, 3 insertions(+) diff --git a/doc/guides/prog_guide/packet_distrib_lib.rst b/doc/guides/prog_guide/packet_distrib_lib.rst index b5bdabb..e0adcaa 100644 --- a/doc/guides/prog_guide/packet_distrib_l

[dpdk-dev] [PATCH v8 15/18] examples/distributor: limit number of Tx rings

Signed-off-by: David Hunt --- examples/distributor/main.c | 23 ++- 1 file changed, 14 insertions(+), 9 deletions(-) diff --git a/examples/distributor/main.c b/examples/distributor/main.c index e9ebe5e..cf2e826 100644 --- a/examples/distributor/main.c +++ b/examples/distribut

[dpdk-dev] [PATCH v8 14/18] examples/distributor: give distributor a core

Signed-off-by: David Hunt --- examples/distributor/main.c | 181 ++-- 1 file changed, 123 insertions(+), 58 deletions(-) diff --git a/examples/distributor/main.c b/examples/distributor/main.c index aeb75a8..e9ebe5e 100644 --- a/examples/distributor/main.c

Re: [dpdk-dev] [PATCH 1/2] net/mlx5: add hardware checksum offload for tunnel packets

Shahaf, See comments below, On Tue, Feb 28, 2017 at 05:53:47PM +0200, Shahaf Shuler wrote: > Prior to this commit Tx checksum offload was supported only for the > inner headers. > This commit adds support for the hardware to compute the checksum for the > outer headers as well. > > The support i

[dpdk-dev] [PATCH v8 13/18] sample: distributor: wait for ports to come up

On some machines, ports take several seconds to come up. This patch causes the app to wait. Signed-off-by: David Hunt --- examples/distributor/main.c | 7 --- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/examples/distributor/main.c b/examples/distributor/main.c index 3657e5d

[dpdk-dev] [PATCH v8 12/18] examples/distributor: allow for extra stats

This will allow us to see what's going on at various stages throughout the sample app, with per-second visibility Signed-off-by: David Hunt --- examples/distributor/main.c | 139 +++- 1 file changed, 123 insertions(+), 16 deletions(-) diff --git a/example

[dpdk-dev] [PATCH v8 08/18] lib: make v20 header file private

Signed-off-by: David Hunt --- lib/librte_distributor/Makefile | 1 - 1 file changed, 1 deletion(-) diff --git a/lib/librte_distributor/Makefile b/lib/librte_distributor/Makefile index a812fe4..2b28eff 100644 --- a/lib/librte_distributor/Makefile +++ b/lib/librte_distributor/Makefile @@ -57,7 +57

[dpdk-dev] [PATCH v8 11/18] test: add perf test for distributor burst mode

Signed-off-by: David Hunt --- test/test/test_distributor_perf.c | 75 ++- 1 file changed, 51 insertions(+), 24 deletions(-) diff --git a/test/test/test_distributor_perf.c b/test/test/test_distributor_perf.c index 1dd326b..732d86d 100644 --- a/test/test/test_d

[dpdk-dev] [PATCH v8 10/18] test: test single and burst distributor API

Signed-off-by: David Hunt --- test/test/test_distributor.c | 116 ++- 1 file changed, 82 insertions(+), 34 deletions(-) diff --git a/test/test/test_distributor.c b/test/test/test_distributor.c index 7a30513..890a852 100644 --- a/test/test/test_distributor.

[dpdk-dev] [PATCH v8 09/18] lib: add symbol versioning to distributor

Also bumped up the ABI version number in the Makefile Signed-off-by: David Hunt --- lib/librte_distributor/Makefile| 2 +- lib/librte_distributor/rte_distributor.c | 8 lib/librte_distributor/rte_distributor_v20.c | 10 ++ lib/librte_distrib

[dpdk-dev] [PATCH v8 07/18] lib: switch distributor over to new API

This is the main switch over between the legacy API and the new burst API. We rename all the functions in rte_distributor.c to remove the _v1705, and we add in _v20 in the rte_distributor_v20.c At the same time, we need the autotests and sample app to compile properly, hence thosie changes are in

[dpdk-dev] [PATCH v8 06/18] test/distributor: extra params for autotests

In the next few patches, we'll want to test old and new API, so here we're allowing different parameters to be passed to the tests, instead of just a distributor struct. Signed-off-by: David Hunt --- test/test/test_distributor.c | 64 +--- 1 file changed,

[dpdk-dev] [PATCH v8 04/18] lib: add new distributor code

This patch includes public header file which will be used once we add in the symbol versioning for v20 and v1705 APIs. Also includes v1702 header file, and code for new burst-capable distributor library. This will be re-named as rte_distributor.h later in the patch-set The new distributor code co

[dpdk-dev] [PATCH v8 05/18] lib: add SIMD flow matching to distributor

Add an optimised version of the in-flight flow matching algorithm using SIMD instructions. This should give up to 1.5x over the scalar versions performance. Falls back to scalar version if SSE4.2 not available Signed-off-by: David Hunt --- lib/librte_distributor/Makefile| 1

[dpdk-dev] [PATCH v8 03/18] lib: add new burst oriented distributor structs

Signed-off-by: David Hunt --- lib/librte_distributor/rte_distributor_private.h | 56 1 file changed, 56 insertions(+) diff --git a/lib/librte_distributor/rte_distributor_private.h b/lib/librte_distributor/rte_distributor_private.h index 6d72f1c..d3a470e 100644 --- a/lib

[dpdk-dev] [PATCH v8 02/18] lib: create private header file

We'll be adding internal implementation definitions in here that are common to both burst and legacy APIs. Signed-off-by: David Hunt --- lib/librte_distributor/rte_distributor_private.h | 136 +++ lib/librte_distributor/rte_distributor_v20.c | 72 +--- 2 files ch

[dpdk-dev] [PATCH v8 0/18] distributor library performance enhancements

This patch aims to improve the throughput of the distributor library. It uses a similar handshake mechanism to the previous version of the library, in that bits are used to indicate when packets are ready to be sent to a worker and ready to be returned from a worker. One main difference is that in

[dpdk-dev] [PATCH v8 01/18] lib: rename legacy distributor lib files

Move files out of the way so that we can replace with new versions of the distributor libtrary. Files are named in such a way as to match the symbol versioning that we will apply for backward ABI compatibility. Signed-off-by: David Hunt --- lib/librte_distributor/Makefile|

[dpdk-dev] [PATCH] net/i40e: fix error log in descriptor done function

It's not queue identifier but a descriptor identifier. Fixes: 4861cde46116 ("i40e: new poll mode driver") Signed-off-by: Olivier Matz --- drivers/net/i40e/i40e_rxtx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c

Re: [dpdk-dev] [PATCH v2] mk: Provide option to set Major ABI version

On Wed, Mar 1, 2017 at 10:34 AM, Christian Ehrhardt wrote: > Downstreams might want to provide different DPDK releases at the same > time to support multiple consumers of DPDK linked against older and newer > sonames. > > Also due to the interdependencies that DPDK libraries can have applications

Re: [dpdk-dev] [PATCH v2 1/1] net/mlx5: add hardware TSO support

Sahaf, Some few remarks below. On Wed, Mar 01, 2017 at 01:11:42PM +0200, Shahaf Shuler wrote: > Implement support for hardware TSO. > > Signed-off-by: Shahaf Shuler > --- > on v2: > * Instead of exposing capability, TSO checks on data path. > * PMD specific parameter to enable TSO. > * diffe

Re: [dpdk-dev] seg fault in InterVM communication using virtio

Hi, We have some doubt regarding virtio front end. As we are running dpdk application inside VM with virtio setup, do we need to use virtio specific functions like virtio_recv_pkts() and virtio_xmit_pkts() to receive and transmit packets, functions for port initialization like virtio_dev_queu

Re: [dpdk-dev] [PATCH v2 11/15] net/avp: packet receive functions

2017-03-01 13:23, Legacy, Allain: > > -Original Message- > > From: Bruce Richardson [mailto:bruce.richard...@intel.com] > > In my experience, checkpatch ignores long lines that are due to error > > messages. Perhaps you need to put the error message on a separate line, > > if other things b

Re: [dpdk-dev] [PATCH v2 08/15] net/avp: device initialization

> -Original Message- > From: Jerin Jacob [mailto:jerin.ja...@caviumnetworks.com] > > +/* Ethernet device validation marker */ #define > RTE_AVP_ETHDEV_MAGIC > > +0x92972862 > > I think, we don't need to add RTE_ for internal flags and PMD APIs etc. Ok, will rename. > > +/* 32-bit MMIO reg

Re: [dpdk-dev] [PATCH v2 02/15] net/avp: public header files

> > +#ifndef RTE_AVP_ALIGNMENT > > +#define RTE_AVP_ALIGNMENT 64 > > I think we use RTE_CACHE_LINE_SIZE here? PPC and ThunderX1 targets are > cache line size of 128B We need this to stay aligned with our host compile environment so we are going to retain this as a local value instead of relying o

Re: [dpdk-dev] [PATCH 4/5] net/vhost: remove limit of vhost TX burst size

Hi, Maxime: > -Original Message- > From: Maxime Coquelin [mailto:maxime.coque...@redhat.com] > Sent: Wednesday, March 1, 2017 5:44 PM > To: Yang, Zhiyong ; dev@dpdk.org > Cc: yuanhan@linux.intel.com > Subject: Re: [PATCH 4/5] net/vhost: remove limit of vhost TX burst size > > > > On

Re: [dpdk-dev] [PATCH v2 11/15] net/avp: packet receive functions

> -Original Message- > From: Bruce Richardson [mailto:bruce.richard...@intel.com] > In my experience, checkpatch ignores long lines that are due to error > messages. Perhaps you need to put the error message on a separate line, > if other things before the message are of significant size. I

  1   2   >