[dpdk-dev] DPDK VF and I40E

DPDK Experts, When I run I40E host PF driver with DPDK VF Driver in VM, I see an issue where packets are received in bunch of 4s. See below But the same problem does not show up when run with DPDK PF testpmd with DPDK VF in VM. Since we would like to have I40e regular driver work with DPDK

[dpdk-dev] [PATCH 0/4] rte_ethdev: cleanups

2015-04-09 14:29, Stephen Hemminger: > A bunch of small (almost trivial) patches to fix style and other > issues in the base Ethernet driver interface code. > > Stephen Hemminger (4): > rte_ethdev: remove extra inline > rte_ethdev: whitespace cleanup > rte_ethdev: make tables const >

[dpdk-dev] Saving packet timestamp when reading from pcap

To test my program and for some other uses I sometimes use vdev (libpcap pmd) to read data from a pcap file. Those tests would be a lot easier if the packet timestamp (which is in the cap) was supplied by DPDK, but alas it is not. So I could access it, I placed it in mbuf's userdata for the

[dpdk-dev] [RFC PATCH 0/6] DPDK support to bifurcated driver

Hi, I can receive packets when I mmaped all pci memory not only rx and tx desc. 2015-04-09 11:43 GMT+08:00 ??? : > Hi Cunming, > I applyed bifurc dirver patches and tested it follow your example. > But I can't received packets with testpmd and l2fwd. > Kernel stack can receive

[dpdk-dev] [PATCH v4 12/12] test/mbuf: verify that cloning a clone works properly

Signed-off-by: Olivier Matz --- app/test/test_mbuf.c | 26 ++ 1 file changed, 26 insertions(+) diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c index 01838c6..b5ae5b7 100644 --- a/app/test/test_mbuf.c +++ b/app/test/test_mbuf.c @@ -324,6 +324,7 @@

[dpdk-dev] [PATCH v4 11/12] test/mbuf: enhance mbuf refcnt test

Check that the data in the cloned mbuf is the same than in the reference mbuf. Check that the reference counter is incremented for each segment. Signed-off-by: Olivier Matz --- app/test/test_mbuf.c | 37 + 1 file changed, 37 insertions(+) diff --git

[dpdk-dev] [PATCH v4 10/12] test/mbuf: rename mc variable in m

It's better to name the mbuf 'm' instead of 'mc' as it's not a clone. Signed-off-by: Olivier Matz --- app/test/test_mbuf.c | 25 - 1 file changed, 12 insertions(+), 13 deletions(-) diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c index 4774263..2614598 100644

[dpdk-dev] [PATCH v4 09/12] mbuf: allow to clone an indirect mbuf

Remove one limitation of rte_pktmbuf_attach(): "mbuf we're attaching to must be direct". Now, when we attach to an indirect mbuf: - copy the all relevant fields (addr, len, offload, ...) as before - get the pointer to the mbuf that embeds the data buffer (direct mbuf), and increase the

[dpdk-dev] [PATCH v4 08/12] mbuf: fix clone support when application uses private mbuf data

Add a new private_size field in mbuf structure that should be initialized at mbuf pool creation. This field contains the size of the application private data in mbufs. Introduce new static inline functions rte_mbuf_from_indirect() and rte_mbuf_to_baddr() to replace the existing macros, which take

[dpdk-dev] [PATCH v4 07/12] apps: use rte_pktmbuf_pool_create to create mbuf pools

When it's possible, use the new helper to create the mbuf pools. Most of the patch is trivial, except for the following files that have some specifics (indirect mbufs): - ip_fragmentation - ip_pipeline - ipv4_multicast - vhost Signed-off-by: Olivier Matz --- app/test-pipeline/init.c

[dpdk-dev] [PATCH v4 06/12] mbuf: introduce a new helper to create a mbuf pool

Add a new wrapper to rte_mempool_create() to simplify the creation of a packet mbuf pool. This wrapper can be used if there is no specific mempool flags, and no specific mbuf or pool constructor function, which is most of the use cases. Signed-off-by: Olivier Matz ---

[dpdk-dev] [PATCH v4 05/12] testpmd: use standard functions to initialize mbufs and mbuf pool

The rte_pktmbuf_pool_init() and rte_pktmbuf_init() functions now support to have a non-hardcoded buffer length. We can remove the specific functions used in testpmd and replace them by the standard ones. Signed-off-by: Olivier Matz --- app/test-pmd/testpmd.c | 74

[dpdk-dev] [PATCH v4 04/12] mbuf: fix rte_pktmbuf_init when mbuf private size is not zero

Allow the user to use the default rte_pktmbuf_init() function even if the mbuf private size is not 0. Signed-off-by: Olivier Matz --- lib/librte_mbuf/rte_mbuf.c | 13 - 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/lib/librte_mbuf/rte_mbuf.c

[dpdk-dev] [PATCH v4 03/12] mbuf: add accessors to get data room size and private size

This code retrieving the pool private area is duplicated in many places, we can use of function for it. Signed-off-by: Olivier Matz --- lib/librte_ether/rte_ethdev.c| 4 +-- lib/librte_mbuf/rte_mbuf.h | 41

[dpdk-dev] [PATCH v4 02/12] examples: always initialize mbuf_pool private area

The mbuf pool private area must always be populated in a mbuf pool. The applications or drivers may expect that for a mbuf pool, the mbuf pool private area (mbuf_data_room_size and mbuf_priv_size) are properly filled. Signed-off-by: Olivier Matz --- examples/ip_fragmentation/main.c | 4 ++--

[dpdk-dev] [PATCH v4 01/12] mbuf: fix mbuf data room size calculation rte_pktmbuf_pool_init

Deduct the mbuf data room size from mempool->elt_size and priv_size, instead of using an hardcoded value that is not related to the real buffer size. To use rte_pktmbuf_pool_init(), the user can either: - give a NULL parameter to rte_pktmbuf_pool_init(): in this case, the private size is

[dpdk-dev] [PATCH v4 00/12] mbuf: enhancements of mbuf clones

The first objective of this series is to fix the support of indirect mbufs when the application reserves a private area in mbufs. It also removes the limitation that rte_pktmbuf_clone() is only allowed on direct (non-cloned) mbufs. The series also contains some enhancements and fixes in the mbuf

[dpdk-dev] QoS Question

Hi DPDK team, The docs on QoS (http://dpdk.org/doc/guides/prog_guide/qos_framework.html# ) describe the traffic class (TC) as follows: 1 - The TCs of the same pipe handled in strict priority order. 2 - Upper limit enforced per TC at the pipe level. 3 - Lower priority TCs able to reuse pipe

[dpdk-dev] [PATCH 18/18] i40e: Minor enhancements in i40e_type.h

To be consistent with base driver release, minor enhancements were added in i40e_type.h, as follows. - 'an_enabled' was removed from 'struct i40e_link_status', as it was useless. - 'requested_speeds' was added to 'struct i40e_link_status' for future use. - 'hw_semaphore_wait' was removed from

[dpdk-dev] [PATCH 17/18] i40e: support of structure and command length check

'I40E_CHECK_STRUCT_LEN' and 'I40E_CHECK_CMD_LENGTH' can be used to check the length validity of structures and commands at compile time. Those checks were added in i40e_adminq_cmd.h which can report issues as early as possible. Signed-off-by: Helin Zhang ---

[dpdk-dev] [PATCH 16/18] i40e: add more virtual channel operations

Added three virtual channel operations, and assigned values to each of them. To be consistent with base driver releases, virtual channel operation of 'I40E_VIRTCHNL_OP_GET_ADDNL_SOL_CONFIG' and its relevant structure were added. Signed-off-by: Helin Zhang ---

[dpdk-dev] [PATCH 15/18] i40e: support of Fortpark FPGA

To support Fortpark FPGA devices, all device IDs and MAC types of FPGA were added. Also, special configurations and processings for FPGA were added. Signed-off-by: Helin Zhang --- lib/librte_pmd_i40e/i40e/i40e_adminq.c| 11 +++ lib/librte_pmd_i40e/i40e/i40e_common.c| 123

[dpdk-dev] [PATCH 14/18] i40e: add new interfaces of AQ commands and relevant

New interfaces were added for AQ commands. They are as follows. - i40e_aq_get_rss_lut() - i40e_aq_set_rss_lut() - i40e_aq_get_rss_key() - i40e_aq_set_rss_key() - i40e_aq_read_nvm_config() - i40e_aq_write_nvm_config() - i40e_aq_set_lldp_mib() - i40e_read_pba_string() Signed-off-by: Helin Zhang

[dpdk-dev] [PATCH 13/18] i40e: replacement of 'i40e_debug_read_register()'

As base driver provides 'i40e_aq_debug_read_register()', the same functional interface of 'i40e_debug_read_register()' can be replaced. Signed-off-by: Helin Zhang --- lib/librte_pmd_i40e/i40e_ethdev.c | 22 ++ 1 file changed, 2 insertions(+), 20 deletions(-) diff --git

[dpdk-dev] [PATCH 12/18] i40e: enhancements of AQ commands and common interfaces

'LE16_TO_CPU' was added in 'i40e_debug_aq()' to support big-endian and little endian correctly. Fix of determing the PF number was added for 'i40e_init_shared_code()', which can solve the issue of PF pass-through. Also enhancements were added for 'i40e_led_get()' and 'i40e_led_set()'.

[dpdk-dev] [PATCH 11/18] i40e: support of building both PF and VF driver together

Macros of PF_DRIVER, VF_DRIVER and INTEGRATED_VF were defined to support building both PF and VF driver together. PF_DRIVER needs to be defined if a build is for PF only, while VF_DRIVER for VF only. PF_DRIVER, VF_DRIVER and INTEGRATED_VF are all needed for building PF and VF driver together.

[dpdk-dev] [PATCH 10/18] i40e: support of firmware build number

Getting and checking firmware build number were added, as firmware supports that. Signed-off-by: Helin Zhang --- lib/librte_pmd_i40e/i40e/i40e_adminq.c| 1 + lib/librte_pmd_i40e/i40e/i40e_adminq.h| 5 +++-- lib/librte_pmd_i40e/i40e/i40e_common.c| 4

[dpdk-dev] [PATCH 09/18] i40e: adminq enhancements

To be more extendable, 'i40e_is_vf()' was introduced to check if the MAC type is VF. Also other minor enhancements were added for 'i40e_init_adminq()'. Signed-off-by: Helin Zhang --- lib/librte_pmd_i40e/i40e/i40e_adminq.c | 31 ++-

[dpdk-dev] [PATCH 08/18] i40e: support NVM read on Fortpark, with minor enhancements

NVM read on Fortpark should be done via AQ command, while others via SRCTL register. So it should check the MAC type first, and then use different functions to do NVM read. Other minor changes were also added for the enhancements of NVM write, checksum calculation. Signed-off-by: Helin Zhang ---

[dpdk-dev] [PATCH 07/18] i40e: rename 'err' to 'perrno'

To be consistent with original base driver, the variable name of 'err' should be renamed to 'perrno'. Signed-off-by: Helin Zhang --- lib/librte_pmd_i40e/i40e/i40e_nvm.c | 118 ++-- 1 file changed, 59 insertions(+), 59 deletions(-) diff --git

[dpdk-dev] [PATCH 06/18] i40e: support of Fortpark device IDs and mac types

Support of Fortpark device IDs and mac types which includes FPGA were added. Signed-off-by: Helin Zhang --- lib/librte_pmd_i40e/i40e/i40e_common.c | 17 + lib/librte_pmd_i40e/i40e/i40e_type.h | 16 2 files changed, 33 insertions(+) diff --git

[dpdk-dev] [PATCH 05/18] i40e: rework of 'i40e_hmc_get_object_va'

To simplify the usage of invoking 'i40e_hmc_get_object_va', input parameters of it were reworked. In addition, minor code style fixes were added as well. Signed-off-by: Helin Zhang --- lib/librte_pmd_i40e/i40e/i40e_lan_hmc.c | 32 +---

[dpdk-dev] [PATCH 04/18] i40e: support of CEE DCBX on recent firmware versions

Recent firmware versions support CEE DCBX, while old firmware versions support IEEE DCBX only. So firmware version needs to be checked during getting DCB configurations. To support CEE DCBX, AQ commands of 'i40e_aq_get_cee_dcb_config' and 'i40e_aq_start_stop_dcbx' were implemented. Signed-off-by:

[dpdk-dev] [PATCH 03/18] i40e: adjustment of register definitions and relevant

More register definitions and their relevant masks are added publically. Also few useless macros are deleted. Signed-off-by: Helin Zhang --- lib/librte_pmd_i40e/i40e/i40e_register.h | 1981 +- lib/librte_pmd_i40e/i40e_ethdev.c|3 - 2 files changed, 1955

[dpdk-dev] [PATCH 02/18] i40e: disable setting phy configuration

There was a known linke issue on 40G ports on NVM version (FVL3E), when setting phy configuration. As a workaround, setting phy configuration should be disabled. The impact is that the link cannot be forcedly configured, which doesn't affect any feature functions. The workaround can be removed

[dpdk-dev] [PATCH 01/18] i40e: copyright update

Copyright is updated. Signed-off-by: Helin Zhang --- lib/librte_pmd_i40e/Makefile | 2 +- lib/librte_pmd_i40e/i40e/i40e_adminq.c | 2 +- lib/librte_pmd_i40e/i40e/i40e_adminq.h | 2 +- lib/librte_pmd_i40e/i40e/i40e_adminq_cmd.h | 2 +- lib/librte_pmd_i40e/i40e/i40e_alloc.h

[dpdk-dev] [PATCH 00/18] i40e base driver update

To support firmware version 'FVL3E', i40e base driver should be updated. It mainly includes the base driver update which contains additional enhancements, fixes, changes for future use, and so on, together with neccessary modifications in i40e Poll Mode Driver. The details are listed as follows.

[dpdk-dev] [RFC PATCH] ethdev: remove old flow director API

It's time to remove this old API. It seems some work is still needed to rely only on eth_ctrl API. At least ixgbe, i40e and testpmd must be fixed. Jingjing, do you think it's possible to remove all these structures from rte_ethdev.h? Thanks --- lib/librte_ether/rte_ethdev.c | 260

[dpdk-dev] [RFC PATCH 1/4] Add example pktdev implementation

On Mon, Apr 20, 2015 at 12:26:43PM +0100, Ananyev, Konstantin wrote: > > > > -Original Message- > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Bruce Richardson > > Sent: Friday, April 17, 2015 4:17 PM > > To: dev at dpdk.org; Wiles, Keith > > Subject: [dpdk-dev] [RFC PATCH

[dpdk-dev] [PATCH] enic: migrating to new fdir api

2015-04-09 14:59, Sujith Sankar: > This patch helps enic migrate to the new flow-director API. > > It takes care of the following. > 1. The change in fdir_filter structure and stats structure > 2. DPDK interface functions in enic_ethdev.c > 3. ENIC driver functions that deal with the VIC

[dpdk-dev] [PATCH] enic: set correct port ID in received mbufs

> >This field is not supposed to contain the RX queue index. Applications can > >rely on it to determine the port a given mbuf comes from. > > > >Signed-off-by: Adrien Mazarguil > > Acked-by: Sujith Sankar Applied, thanks

[dpdk-dev] [PATCH] pci: make rte_pci_probe void

2015-04-14 10:55, Stephen Hemminger: > Since rte_pci_probe always returns 0 or exits via rte_exit() > there is no point in having it return a value. > > Just make it void > > Signed-off-by: Stephen Hemminger Seems partially superseded by this patch:

[dpdk-dev] [PATCH 2/2] eal: pci probe and adjust_config should be void

2015-04-17 08:35, Stephen Hemminger: > This functions always returned 0 and therefore should be void. > > Signed-off-by: Stephen Hemminger > --- > lib/librte_eal/common/eal_common_options.c | 3 +-- > lib/librte_eal/common/eal_common_pci.c | 7 ++- > lib/librte_eal/common/eal_options.h

[dpdk-dev] [PATCH 7/7] hv: add kernel patch

From: Stephen Hemminger For users using non latest kernels, put kernel patch in for them to use. Signed-off-by: Stephen Hemminger --- .../linuxapp/hv_uio/vmbus-get-pages.patch | 55 ++ 1 file changed, 55 insertions(+) create mode 100644

[dpdk-dev] [PATCH 6/7] hv: enable driver in common config

From: Stephen Hemminger Add hyperv driver config to enable it. Signed-off-by: Stephen Hemminger --- config/common_linuxapp | 9 + 1 file changed, 9 insertions(+) diff --git a/config/common_linuxapp b/config/common_linuxapp index 0078dc9..58cc352 100644 ---

[dpdk-dev] [PATCH 5/7] hv: poll mode driver

From: Stephen Hemminger This is new Poll Mode driver for using hyper-v virtual network interface. Signed-off-by: Stas Egorov Signed-off-by: Stephen Hemminger --- lib/Makefile |1 + lib/librte_pmd_hyperv/Makefile| 28 +

[dpdk-dev] [PATCH 4/7] hv: uio driver

From: Stephen Hemminger Add new UIO driver in kernel to support DPDK Poll Mode Driver. Signed-off-by: Stas Egorov Signed-off-by: Stephen Hemminger --- lib/librte_eal/linuxapp/Makefile| 3 + lib/librte_eal/linuxapp/hv_uio/Makefile | 57 ++

[dpdk-dev] [PATCH 3/7] hv: add basic vmbus support

The hyper-v device driver forces the base EAL code to change to support multiple bus types. This is done changing the pci_device in ether driver to a generic union. As much as possible this is done in a backwards source compatiable way. It will break ABI for device drivers. Signed-off-by:

[dpdk-dev] [PATCH 2/7] pmd: change drivers initialization for pci

From: Stephen Hemminger The change to generic ether device structure to support multiple bus types requires a change to all existing PMD but only in the initialization (and the change is backwards compatiable). Signed-off-by: Stephen Hemminger ---

[dpdk-dev] [PATCH 1/7] ether: add function to query for link state interrupt

From: Stephen Hemminger Allow application to query whether link state will work. This is also part of abstracting dependency on PCI. Signed-off-by: Stephen Hemminger --- lib/librte_ether/rte_ethdev.c | 14 ++ lib/librte_ether/rte_ethdev.h | 12 2

[dpdk-dev] [PATCH 0/7] Hyper-V Poll Mode Driver

This is revised version of earlier patches to support Hyper-V poll mode driver. The main changes were to rebase onto DPDK 2.0 with a few other style cleanups to keep checkpatch at bay. Stephen Hemminger (7): ether: add function to query for link state interrupt pmd: change drivers

[dpdk-dev] [PATCH v3 2/2] use simple zero initializers

> > To initialize a structure with zeros, one field was explicitly set > > to avoid "missing initializer" bug with old GCC (e.g. 4.4). > > This warning is now disabled (commit ) for old versions of GCC, > > so the workarounds may be removed. > > > > These initializers should not be needed for

[dpdk-dev] [PATCH v3 1/2] mk: fix build with gcc 4.4 and clang

> > With GCC 4.4.7 from CentOS 6.5, the following errors arise: > > > > lib/librte_pmd_ixgbe/ixgbe_rxtx.c: In function 'ixgbe_dev_rx_queue_setup': > > lib/librte_pmd_ixgbe/ixgbe_rxtx.c:2509: error: missing initializer > > lib/librte_pmd_ixgbe/ixgbe_rxtx.c:2509: error: (near initialization for > >

[dpdk-dev] cost of reading tsc register

Hi, We were doing some code optimizations , running DPDK based applications, and chanced upon the rte_rdtsc function [ to read tsc timestamp register value ] consuming cpu cycles of the order of 100clock cycles with a delta of upto 40cycles at times [ 60-140 cycles] We are actually building up

[dpdk-dev] [RFC PATCH] ethdev: remove old flow director API

> -Original Message- > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com] > Sent: Monday, April 20, 2015 10:12 PM > To: Wu, Jingjing > Cc: dev at dpdk.org > Subject: [RFC PATCH] ethdev: remove old flow director API > > It's time to remove this old API. > It seems some work is

[dpdk-dev] [PATCH v2] Clean up rte_memcpy.h file

Remove unnecessary type casting in functions. Tested on Ubuntu (14.04 x86_64) with "make test". "make test" results match the results with baseline. "Memcpy perf" results match the results with baseline. Signed-off-by: Ravi Kerur --- .../common/include/arch/x86/rte_memcpy.h | 340

[dpdk-dev] [PATCH v2] Clean up rte_memcpy.h

This version contains changes for removing unnecessary typecasting only. Backing out remaining changes i.e. loop-unrolling. Though loop-unrolling makes sense from more space/less time perspective, code generated by GCC 4.8.2 with "gcc -O3 -mavx -s" and "gcc -O3 -m64 -s" for loop of 2, 4 and 8

[dpdk-dev] [RFC PATCH 0/4] pktdev

On 4/20/15, 8:19 AM, "Wiles, Keith" wrote: > > >From: Marc Sune mailto:marc.sune at bisdn.de>> >Date: Monday, April 20, 2015 at 1:51 AM >To: Keith Wiles mailto:keith.wiles at intel.com>>, >"dev at dpdk.org" mailto:dev at >dpdk.org>> >Subject: Re: [dpdk-dev] [RFC PATCH

[dpdk-dev] DCA

Hi, I would like to ask if there is any reason why DPDK doesn't have support for DCA feature? thanks, vlad

[dpdk-dev] [RFC PATCH] ethdev: remove old flow director API

On Mon, Apr 20, 2015 at 04:11:43PM +0200, Thomas Monjalon wrote: > It's time to remove this old API. > It seems some work is still needed to rely only on eth_ctrl API. > At least ixgbe, i40e and testpmd must be fixed. > Jingjing, do you think it's possible to remove all these structures > from

[dpdk-dev] [PATCH v2 4/4] bond mode 4: tests for external state machine

From: Eric Kinzie This adds test cases for exercising the external state machine API to the mode 4 autotest. Signed-off-by: Eric Kinzie --- app/test/test_link_bonding_mode4.c | 210 ++-- 1 file changed, 201 insertions(+), 9

[dpdk-dev] [PATCH v2 3/4] bond mode 4: allow external state machine

From: Eric Kinzie Provide functions to allow an external 802.3ad state machine to transmit and recieve LACPDUs and to set the collection/distribution flags on slave interfaces. Signed-off-by: Eric Kinzie --- lib/librte_pmd_bond/rte_eth_bond_8023ad.c | 173

[dpdk-dev] [PATCH v2 2/4] bond mode 4: do not ignore multicast

From: Eric Kinzie The bonding PMD in mode 4 puts all enslaved interfaces into promiscuous mode in order to receive LACPDUs and must filter unwanted packets after the traffic has been "collected". Allow broadcast and multicast through so that ARP and IPv6 neighbor discovery

[dpdk-dev] [PATCH v2 1/4] bond mode 4: copy entire config structure

From: Eric Kinzie Copy all needed fields from the mode8023ad_private structure in bond_mode_8023ad_conf_get(). This help ensure that a subsequent call to rte_eth_bond_8023ad_setup() is not passed uninitialized data that would result in either incorrect behavior or a

[dpdk-dev] [PATCH v2 0/4] bonding corrections and additions

This patchset makes a couple of small corrections to the bonding driver and introduces the ability to use an external state machine for mode 4 operation. Changes in v2: . eliminate external_sm field in 802.3ad configuration (rte_eth_bond_8023ad_conf). . stop bonding device before

[dpdk-dev] DCA

On Mon, Apr 20, 2015 at 01:07:59PM +0300, Vlad Zolotarov wrote: > Hi, > I would like to ask if there is any reason why DPDK doesn't have support for > DCA feature? > > thanks, > vlad With modern platforms with DDIO the data written by the NIC automatically goes into the cache of the CPU without

[dpdk-dev] [RFC PATCH 0/4] pktdev

On Mon, Apr 20, 2015 at 08:51:26AM +0200, Marc Sune wrote: > > > On 17/04/15 21:50, Wiles, Keith wrote: > >Hi Marc and Bruce, > > Hi Keith, Bruce, > > > > >On 4/17/15, 1:49 PM, "Marc Sune" wrote: > What I was proposing is to try to add the minimum common shared state in > order to properly

[dpdk-dev] [RFC PATCH 1/4] Add example pktdev implementation

> -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Bruce Richardson > Sent: Friday, April 17, 2015 4:17 PM > To: dev at dpdk.org; Wiles, Keith > Subject: [dpdk-dev] [RFC PATCH 1/4] Add example pktdev implementation > > This commit demonstrates what a minimal

[dpdk-dev] [RFC PATCH 3/4] add support for a ring to be a pktdev

Hi Bruce, > -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Bruce Richardson > Sent: Friday, April 17, 2015 4:17 PM > To: dev at dpdk.org; Wiles, Keith > Subject: [dpdk-dev] [RFC PATCH 3/4] add support for a ring to be a pktdev > > Add a new public API

[dpdk-dev] cost of reading tsc register

On Mon, Apr 20, 2015 at 02:37:53PM +, Ravi Kumar Iyer wrote: > We were doing some code optimizations , running DPDK based applications, and > chanced upon the rte_rdtsc function [ to read tsc timestamp register value ] > consuming cpu cycles of the order of 100clock cycles with a delta of

[dpdk-dev] [RFC PATCH 0/4] pktdev

On 17/04/15 21:50, Wiles, Keith wrote: > Hi Marc and Bruce, Hi Keith, Bruce, > > On 4/17/15, 1:49 PM, "Marc Sune" wrote: > >> >> On 17/04/15 17:16, Bruce Richardson wrote: >>> Hi all, >>> >>> to continue this discussion a bit more, here is my, slightly different, >>> slant >>> on what a

[dpdk-dev] cost of reading tsc register

On Mon, 20 Apr 2015 14:37:53 + Ravi Kumar Iyer wrote: > Hi, > We were doing some code optimizations , running DPDK based applications, and > chanced upon the rte_rdtsc function [ to read tsc timestamp register value ] > consuming cpu cycles of the order of 100clock cycles with a delta of

[dpdk-dev] [PATCH RFC 00/10] Add a VXLAN sample

Any comments on this RFC patch set ? > -Original Message- > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jijiang Liu > Sent: Thursday, April 16, 2015 11:56 AM > To: dev at dpdk.org; Gilmore, Walter E; Long, Thomas > Subject: [dpdk-dev] [PATCH RFC 00/10] Add a VXLAN sample > >