[dpdk-dev] [PATCH 0/2] extend RSS offload types

From: Simei Su [PATCH 1/2] ethdev: add several bits for extending rss offload types. [PATCH 2/2] app/testpmd: add cmdline support for extending rss types. Simei Su (2): ethdev: extend RSS offload types app/testpmd: add RSS offload types extending support app/test-pmd/config.c | 6

[dpdk-dev] [PATCH 1/2] ethdev: extend RSS offload types

From: Simei Su This patch cover two aspects: (1)decouple RTE_ETH_FLOW_* and ETH_RSS_*. Because both serve different purposes. (2)reserve several bits as input set selection from bottom of the 64 bits. It is combined with exisitingi ETH_RSS_* to represent rss types. fo

[dpdk-dev] [PATCH 2/2] app/testpmd: add RSS offload types extending support

From: Simei Su This patch adds cmdline support for extended rss types configuration. Signed-off-by: Simei Su --- app/test-pmd/config.c | 6 ++ 1 file changed, 6 insertions(+) diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c index 1a5a5c1..b95bd43 100644 --- a/app/test-pmd/config

[dpdk-dev] 19.11 Marvell Roadmap

1) eBPF arm64 JIT support 2) Enhance UT cases for eBPF verification 3) Add new l2fwd-event eventdev application 4) Update l3fwd application to support event mode 5) Update ipsec-secgw application to support event mode 6) Add build infrastructure support for UBSan 7) ethdev enhancements a) As

Re: [dpdk-dev] [patch v3] doc: announce API change in ethdev offload flags

> -Original Message- > From: Ananyev, Konstantin > Sent: Thursday, August 8, 2019 10:24 PM > To: Jerin Jacob Kollanukkaran ; Pavan Nikhilesh > Bhagavatula ; step...@networkplumber.org; > arybche...@solarflare.com; hemant.agra...@nxp.com; > tho...@monjalon.net; Yigit, Ferruh ; Richardson, >

Re: [dpdk-dev] [v2] net/i40e: fix vf runtime queues rss config

> -Original Message- > From: Xing, Beilei > Sent: Friday, August 9, 2019 10:44 AM > To: Zhang, Xiao ; dev@dpdk.org > Cc: Zhang, Qi Z ; sta...@dpdk.org > Subject: RE: [v2] net/i40e: fix vf runtime queues rss config > > > > -Original Message- > > From: Zhang, Xiao > > Sent: Frida

Re: [dpdk-dev] which is the code base when submitting patch around end of September

Hello Ye, Thanks for your quick response. I understand the basis. I don't have much intention to push proposal to 19.11 right now. Thanks again. BR, Hideyuki Yamashita NTT TechnoCross > Hi, > > On 08/09, Hideyuki Yamashita wrote: > >Hello Experts, > > > > > >Very basic question to create patc

Re: [dpdk-dev] [v2] net/i40e: fix vf runtime queues rss config

> -Original Message- > From: Zhang, Xiao > Sent: Friday, August 9, 2019 12:41 AM > To: dev@dpdk.org > Cc: Xing, Beilei ; Zhang, Qi Z ; > Zhang, Xiao ; sta...@dpdk.org > Subject: [v2] net/i40e: fix vf runtime queues rss config > > I40evf queue can not work properly with kernel pf driver.

Re: [dpdk-dev] which is the code base when submitting patch around end of September

Hi, On 08/09, Hideyuki Yamashita wrote: >Hello Experts, > > >Very basic question to create patch. > >I am planning to post patch set which introduce new functionality (not >bug fix ). >Note that I will post the patch set on end of September. If your patch is targeted for 19.11 release, you need t

[dpdk-dev] which is the code base when submitting patch around end of September

Hello Experts, Very basic question to create patch. I am planning to post patch set which introduce new functionality (not bug fix ). Note that I will post the patch set on end of September. Q1. If I post patch set to dev-ml, which is the code base? Source code which is retrieved from "git clo

[dpdk-dev] [PATCH] test/interrupt: account for race with callback

Because the eal interrupt framework can race when invoking the callback and a separate unregister call, the test needs to accommodate the chance that the two collide. Do this by checking the return value of unregister against the race-condition flag (EAGAIN). Fixes: f1a6c22424ce ("app/test: updat

Re: [dpdk-dev] [patch v3] doc: announce API change in ethdev offload flags

> > > > > > > Add new offload flags ``DEV_RX_OFFLOAD_PTYPE``, > > > > > > ``DEV_RX_OFFLOAD_RSS`` > > > > > > > and ``DEV_RX_OFFLOAD_FLOW_MARK``. > > > > > > > > > > > > > > Signed-off-by: Pavan Nikhilesh > > > > > > > Acked-by: Andrew Rybchenko > > > > > > > Acked-by: Jerin Jacob > > > > > > > -

Re: [dpdk-dev] [PATCH 2/2] doc: announce new mbuf field for LRO

> Hi > From: Ananyev, Konstantin > > Hi Matan, > > > > > > > > > > > > > > > > > > > > The API breakage is because the ``tso_segsz`` field was > > > > > > > > > documented for LRO. > > > > > > > > > > > > > > > > > > The ``tso_segsz`` field in mbuf indicates the size of each > > > > > > > > > se

[dpdk-dev] [PATCH v2] net/ipn3ke: setup MTU when HW init

set up mtu to the minimun in tx mtu, rx mtu and IPN3KE_MAC_FRAME_SIZE_MAX. Signed-off-by: Andy Pei --- v2: * modify low bound and upper bound. drivers/net/ipn3ke/ipn3ke_ethdev.c | 97 ++ drivers/net/ipn3ke/ipn3ke_ethdev.h | 19 2 files changed, 116 i

Re: [dpdk-dev] [PATCH v2] doc: announce lcore_config symbol removal

08/08/2019 11:31, David Marchand: > New accessors have been introduced to provide the hidden information. > This symbol can now be kept internal. > > Signed-off-by: David Marchand > Acked-by: Stephen Hemminger > Acked-by: Bruce Richardson > Acked-by: Jerin Jacob > Acked-by: Andrew Rybchenko A

Re: [dpdk-dev] [PATCH v2] doc: announce malloc virt2phys symbol removal

08/08/2019 11:32, David Marchand: > This symbol has been deprecated for quite some time. > Let's drop it as soon as possible. > > Signed-off-by: David Marchand > Acked-by: Anatoly Burakov > Acked-by: Jerin Jacob > Acked-by: Andrew Rybchenko > Acked-by: Konstantin Ananyev > --- > Changelog sin

Re: [dpdk-dev] [PATCH v3] usertools: fix input handling in telemetry script

08/08/2019 16:37, Andrius Sirvys: > This commit removes the unnecesarry ast.literal_eval() function call > from the input handling, which now relies just on raw_input() to get > its input. > > Fixes: 53f293c9a783 ("usertools: replace unsafe input function") > Cc: andrius.sir...@intel.com > > Sign

Re: [dpdk-dev] [PATCH v3] usertools: fix input handling in telemetry script

> -Original Message- > From: Sirvys, Andrius > Sent: Thursday, August 8, 2019 3:38 PM > To: Laatz, Kevin > Cc: dev@dpdk.org; sta...@dpdk.org; Van Haaren, Harry > ; Ferriter, Cian ; > Lipiec, Herakliusz ; Sirvys, Andrius > > Subject: [PATCH v3] usertools: fix input handling in telemetry sc

Re: [dpdk-dev] [dpdk-stable] [PATCH v2] eal: change max hugepage sizes to 4 for ARM platforms

08/08/2019 14:37, David Marchand: > On Thu, Aug 8, 2019 at 2:27 PM Gagandeep Singh wrote: > > > > ARM is supporting maximum 4 hugepage sizes (64K, 2M, 32M > > and 1G) when granule is 4KB since very long and DPDK > > support maximum 3 hugepage sizes. > > > > With all 4 hugepage sizes enabled, appli

[dpdk-dev] [dpdk-announce] DPDK Summit North America: CFP Now Open! (Nov. 12-13)

Hello DPDK Community, The CFP for DPDK Summit North America 2019, happening November 12-13 in Mountain View, CA, is now open. As with previous DPDK Summits, we are looking for this year’s agenda to cover the latest developments to the DPD

Re: [dpdk-dev] [PATCH] eal: change max hugepage sizes to 4

Hello, On Thu, Aug 08, 2019 at 10:37:20AM +, Hemant Agrawal wrote: > Hi David, > > > > VPP usages in-memory option. So, VPP on ARM with kernel 5.2 wont' work > > without this patch. > > > > > > > I have been looking at the changes in the linux kernel. > > Can you pinpoint at the commit that

[dpdk-dev] [PATCH v3] usertools: fix input handling in telemetry script

This commit removes the unnecesarry ast.literal_eval() function call from the input handling, which now relies just on raw_input() to get its input. Fixes: 53f293c9a783 ("usertools: replace unsafe input function") Cc: andrius.sir...@intel.com Signed-off-by: Andrius Sirvys --- v2: Removed unneces

Re: [dpdk-dev] [PATCH 19.11 V3 00/12] Miscellaneous fixes

08/08/2019 15:19, Thierry Herbelot: > On 8/8/19 10:22 AM, Thierry Herbelot wrote: > > This series of patches includes fixes for issues seen with > > 6WIND fast path, built on DPDK. > > The patches are in 6WIND version of DPDK, and should have been > > upstreamed a long time ago. > > In fact, I wil

Re: [dpdk-dev] [PATCH] ci: exit setup on any error

David Marchand writes: > -e is preferrable so that we can catch errors in the middle of this > script. > An example is this Travis job [1] that should have errored at the meson > install step rather than go to the build step. > > Adding debug mode as it can help post-mortem. > > 1: https://travis

Re: [dpdk-dev] [PATCH] doc: inline data settings affect mlx5 Tx queue max size

08/08/2019 13:47, Viacheslav Ovsiienko: > Introduces the possible limitations on maximal Tx queue > size in descriptors if Tx inline data are enabled. > > Signed-off-by: Viacheslav Ovsiienko Applied, thanks

[dpdk-dev] [PATCH v4 0/2] add unit test cases for mbuf library

This patchset contains unit testcases added to increase the functional and decision coverage for the library functions defined in rte_mbuf.h and rte_mbuf.c 1/2: unit test cases added for rte_mbuf.h 2/2: unit test cases added for rte_mbuf.c Patch 2/2 depends on 1/2 Signed-off-by: Lavanya Govindar

[dpdk-dev] [PATCH v4 1/2] app/test: add unit test cases for mbuf library APIs

Added new unit test cases to cover the below functions defined in rte_mbuf.h rte_validate_tx_offload, rte_pktmbuf_alloc_bulk, rte_pktmbuf_read, rte_pktmbuf_ext_shinfo_init_helper, rte_pktmbuf_attach_extbuf, rte_mbuf_ext_refcnt_read, rte_mbuf_ext_refcnt_update, rte_mbuf_ext_refcnt_set, rte_pktmbuf_d

[dpdk-dev] [PATCH v4 2/2] app/test: add unit test cases to mbuf

From: Pallantla Poornima Added UT for the below four functions in test_mbuf.c rte_get_rx_ol_flag_list rte_get_tx_ol_flag_list rte_get_rx_ol_flag_name rte_get_tx_ol_flag_name Signed-off-by: Pallantla Poornima --- app/test/test_mbuf.c | 260 +++ 1 file cha

Re: [dpdk-dev] [PATCH 19.11 V3 00/12] Miscellaneous fixes

On 8/8/19 10:22 AM, Thierry Herbelot wrote: This series of patches includes fixes for issues seen with 6WIND fast path, built on DPDK. The patches are in 6WIND version of DPDK, and should have been upstreamed a long time ago. In fact, I will re-submit another series of patches, as some are not

Re: [dpdk-dev] [PATCH] ci: exit setup on any error

On Thu, Aug 8, 2019 at 8:22 AM David Marchand wrote: > > -e is preferrable so that we can catch errors in the middle of this > script. > An example is this Travis job [1] that should have errored at the meson > install step rather than go to the build step. > > Adding debug mode as it can help pos

Re: [dpdk-dev] [PATCH] eal: change max hugepage sizes to 4

On Thu, Aug 8, 2019 at 2:29 PM Steve Capper wrote: > > Hello, > > On Thu, Aug 08, 2019 at 10:37:20AM +, Hemant Agrawal wrote: > > Hi David, > > > > > > VPP usages in-memory option. So, VPP on ARM with kernel 5.2 wont' work > > > without this patch. > > > > > > > > > > I have been looking at th

Re: [dpdk-dev] [dpdk-stable] [PATCH v2] eal: change max hugepage sizes to 4 for ARM platforms

On Thu, Aug 8, 2019 at 2:27 PM Gagandeep Singh wrote: > > ARM is supporting maximum 4 hugepage sizes (64K, 2M, 32M > and 1G) when granule is 4KB since very long and DPDK > support maximum 3 hugepage sizes. > > With all 4 hugepage sizes enabled, applications and some > stacks like VPP which are wor

[dpdk-dev] [PATCH] eal: added new api to only enqueue a packet in tx buffer

This api is similar like api `rte_eth_tx_buffer` except it does not attempt to flush the buffer in case buffer is full. The advantage is that, this api does not need port id and queue id. In case port id and queue id are shared within threads then application can not buffer a packet until it gets a

[dpdk-dev] [PATCH v2] eal: change max hugepage sizes to 4 for ARM platforms

ARM is supporting maximum 4 hugepage sizes (64K, 2M, 32M and 1G) when granule is 4KB since very long and DPDK support maximum 3 hugepage sizes. With all 4 hugepage sizes enabled, applications and some stacks like VPP which are working over DPDK and using "in-memory" eal option, or using separate m

[dpdk-dev] [PATCH] ci: exit setup on any error

-e is preferrable so that we can catch errors in the middle of this script. An example is this Travis job [1] that should have errored at the meson install step rather than go to the build step. Adding debug mode as it can help post-mortem. 1: https://travis-ci.com/DPDK/dpdk/jobs/223511683 Signe

[dpdk-dev] [PATCH] eal: added new api to only enqueue a packet in tx buffer

This api is similar like api `rte_eth_tx_buffer` except it does not attempt to flush the buffer in case buffer is full. The advantage is that, this api does not need port id and queue id. In case port id and queue id are shared within threads then application can not buffer a packet until it gets a

Re: [dpdk-dev] [PATCH v2] doc: announce cpuflags check symbol removal

> > This symbol has been deprecated for quite some time. > > Let's drop it as soon as possible. > > > > Signed-off-by: David Marchand > > Acked-by: Aaron Conole > > Acked-by: Konstantin Ananyev > > Acked-by: Andrew Rybchenko > Acked-by: Jerin Jacob Acked-by: Thomas Monjalon > > +* eal: The

[dpdk-dev] [PATCH] doc: inline data settings affect mlx5 Tx queue max size

Introduces the possible limitations on maximal Tx queue size in descriptors if Tx inline data are enabled. Signed-off-by: Viacheslav Ovsiienko --- doc/guides/nics/mlx5.rst | 10 ++ 1 file changed, 10 insertions(+) diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst index d6

[dpdk-dev] [PATCH] net/mlx5: fix missing ptype for IP-in-IP

The hw ptype information is missed for IP-in-IP tunnel. It should be RTE_PTYPE_TUNNEL_IP ptype. Fixes: 5e33bebdd8d3 ("net/mlx5: support IP-in-IP tunnel") Signed-off-by: Xiaoyu Min --- drivers/net/mlx5/mlx5_flow.c | 8 drivers/net/mlx5/mlx5_rxtx.h | 2 +- 2 files changed, 9 insertions(+

Re: [dpdk-dev] [dpdk-stable] [PATCH v3] net/ixgbe: fix offset value check

08/08/2019 02:37, Congwen Zhang: > The "and" condition offset == 0 && offset == NVM_INVALID_PTR > can never be true. > > Fixes: cf3af5aa56c9 ("net/ixgbe/base: add functions to get version info") > Cc: sta...@dpdk.org > > Signed-off-by: Congwen Zhang > Acked-by: Xiaolong Ye Applied, thanks No

Re: [dpdk-dev] [dpdk-stable] [PATCH] test/eal: do not scan PCI devices for memory tests

On Mon, Aug 5, 2019 at 12:20 PM Thomas Monjalon wrote: > 03/08/2019 11:51, David Marchand: > > On Fri, Aug 2, 2019 at 10:57 PM Thomas Monjalon wrote: > > > > > > 01/08/2019 14:27, David Marchand: > > > > The memory tests currently check that, for normal mode (not legacy > > > > mode), > > > > th

[dpdk-dev] [PATCH] eal: added new api to only enqueue a packet in tx buffer

This api is similar like api `rte_eth_tx_buffer` except it does not attempt to flush the buffer in case buffer is full. The advantage is that, this api does not need port id and queue id. In case port id and queue id are shared within threads then application can not buffer a packet until it gets a

Re: [dpdk-dev] [PATCH 2/2] doc: announce new mbuf field for LRO

Hi From: Ananyev, Konstantin > Hi Matan, > > > > > > > > > > > > > > > > > The API breakage is because the ``tso_segsz`` field was > > > > > > > > documented for LRO. > > > > > > > > > > > > > > > > The ``tso_segsz`` field in mbuf indicates the size of each > > > > > > > > segment in the LRO pac

Re: [dpdk-dev] [patch v3] doc: announce API change in ethdev offload flags

08/08/2019 12:59, Jerin Jacob Kollanukkaran: > From: Ananyev, Konstantin > > From: Jerin Jacob Kollanukkaran [mailto:jer...@marvell.com] > > > From: Ananyev, Konstantin > > > > > > > From: Pavan Nikhilesh > > > > > > One question about DEV_RX_OFFLOAD_PTYPE: > > > > > > Does it mean that new ol_f

Re: [dpdk-dev] [patch v3] doc: announce API change in ethdev offload flags

> -Original Message- > From: Ananyev, Konstantin > Sent: Thursday, August 8, 2019 4:04 PM > To: Jerin Jacob Kollanukkaran ; Pavan Nikhilesh > Bhagavatula ; step...@networkplumber.org; > arybche...@solarflare.com; hemant.agra...@nxp.com; > tho...@monjalon.net; Yigit, Ferruh ; Richardson, >

Re: [dpdk-dev] [PATCH 2/2] doc: announce new mbuf field for LRO

Hi Matan, > > > > > > > > > > > > > > The API breakage is because the ``tso_segsz`` field was > > > > > > > documented for LRO. > > > > > > > > > > > > > > The ``tso_segsz`` field in mbuf indicates the size of each > > > > > > > segment in the LRO packet in Rx path and should be provided by > > >

Re: [dpdk-dev] [PATCH] eal: change max hugepage sizes to 4

Hi David, > > VPP usages in-memory option. So, VPP on ARM with kernel 5.2 wont' work > without this patch. > > > > I have been looking at the changes in the linux kernel. > Can you pinpoint at the commit that changed this in 5.2? > > I can see a change in the code, but in 5.0, or maybe something

Re: [dpdk-dev] [PATCH 19.11 V3 02/12] ethdev: fix description of tx descriptor status

On 8/8/19 11:22 AM, Thierry Herbelot wrote: From: Olivier Matz The API comment of rte_eth_tx_descriptor_status() was incorrect. The reference descriptor (when offset = 0) is not where the next packet will be sent, but where the latest packet has been enqueued. Fixes: 52f5cdd2e897 ("ethdev: add

Re: [dpdk-dev] [patch v3] doc: announce API change in ethdev offload flags

> -Original Message- > From: Jerin Jacob Kollanukkaran [mailto:jer...@marvell.com] > Sent: Thursday, August 8, 2019 11:23 AM > To: Ananyev, Konstantin ; Pavan Nikhilesh > Bhagavatula ; > step...@networkplumber.org; arybche...@solarflare.com; > hemant.agra...@nxp.com; tho...@monjalon.ne

Re: [dpdk-dev] [patch v3] doc: announce API change in ethdev offload flags

> -Original Message- > From: Ananyev, Konstantin > Sent: Thursday, August 8, 2019 3:39 PM > To: Jerin Jacob Kollanukkaran ; Pavan Nikhilesh > Bhagavatula ; step...@networkplumber.org; > arybche...@solarflare.com; hemant.agra...@nxp.com; > tho...@monjalon.net; Yigit, Ferruh ; Richardson, >

Re: [dpdk-dev] [PATCH] test/mcslock: wait for lcore completion

08/08/2019 05:44, Phil Yang (Arm Technology China): > From: Aaron Conole > > > > It's possible that the mcsunlock occurs before the test_mcslock_try has > > a chance to execute, which will result in the trylock being successful, > > making the test case fail. Fix this by waiting until all lcores

Re: [dpdk-dev] [PATCH 2/2] doc: announce new mbuf field for LRO

From: Ananyev, Konstantin > > Hi Konstantin > > > > From: Ananyev, Konstantin > > > Sent: Wednesday, August 7, 2019 1:18 PM > > > To: Matan Azrad ; dev@dpdk.org > > > Cc: Thomas Monjalon ; Yigit, Ferruh > > > ; Andrew Rybchenko > > > ; Olivier Matz > > > Subject: RE: [PATCH 2/2] doc: announce n

Re: [dpdk-dev] [dpdk-stable] [PATCH] app/testpmd: fix latencystats deinit on signal

08/08/2019 10:54, Jerin Jacob Kollanukkaran: > > From: Amit Gupta > > > > On receiving signal, testpmd showing warning as > > "LATENCY_STATS: failed to remove Rx/Tx callback" > > because rte_latencystats_uninit is called without checking if latencystats > > is > > enabled or not. > > > > After

Re: [dpdk-dev] [patch v3] doc: announce API change in ethdev offload flags

Hi Jerin, > > > > Hi guys, > > > > > > > > From: Pavan Nikhilesh > > > > > > Add new offload flags ``DEV_RX_OFFLOAD_PTYPE``, > > ``DEV_RX_OFFLOAD_RSS`` > > > and ``DEV_RX_OFFLOAD_FLOW_MARK``. > > > > > > Signed-off-by: Pavan Nikhilesh > > > Acked-by: Andrew Rybchenko > > > Acked-by: Jerin Jacob

Re: [dpdk-dev] [PATCH v2] doc: announce cpuflags check symbol removal

> -Original Message- > From: dev On Behalf Of David Marchand > Sent: Thursday, August 8, 2019 3:01 PM > To: dev@dpdk.org > Cc: tho...@monjalon.net; acon...@redhat.com; > konstantin.anan...@intel.com; arybche...@solarflare.com > Subject: [dpdk-dev] [PATCH v2] doc: announce cpuflags check sy

Re: [dpdk-dev] [patch v3] doc: announce API change in ethdev offload flags

> -Original Message- > From: Ananyev, Konstantin > Sent: Thursday, August 8, 2019 2:53 PM > To: Pavan Nikhilesh Bhagavatula ; Jerin Jacob > Kollanukkaran ; step...@networkplumber.org; > arybche...@solarflare.com; hemant.agra...@nxp.com; > tho...@monjalon.net; Yigit, Ferruh ; Richardson, >

[dpdk-dev] [PATCH v2] doc: announce lcore_config symbol removal

New accessors have been introduced to provide the hidden information. This symbol can now be kept internal. Signed-off-by: David Marchand Acked-by: Stephen Hemminger Acked-by: Bruce Richardson Acked-by: Jerin Jacob Acked-by: Andrew Rybchenko --- Changelog since v1: - added acks, - fixed littl

[dpdk-dev] [PATCH v2] doc: announce malloc virt2phys symbol removal

This symbol has been deprecated for quite some time. Let's drop it as soon as possible. Signed-off-by: David Marchand Acked-by: Anatoly Burakov Acked-by: Jerin Jacob Acked-by: Andrew Rybchenko Acked-by: Konstantin Ananyev --- Changelog since v1: - added acks, - removed mention of 19.11 as the

[dpdk-dev] [PATCH v2] doc: announce cpuflags check symbol removal

This symbol has been deprecated for quite some time. Let's drop it as soon as possible. Signed-off-by: David Marchand Acked-by: Aaron Conole Acked-by: Konstantin Ananyev Acked-by: Andrew Rybchenko --- Changelog since v1: - added acks, - removed mention of 19.11 as the target release, --- doc

Re: [dpdk-dev] [patch v3] doc: announce API change in ethdev offload flags

Hi guys, > > From: Pavan Nikhilesh > > Add new offload flags ``DEV_RX_OFFLOAD_PTYPE``, ``DEV_RX_OFFLOAD_RSS`` > and ``DEV_RX_OFFLOAD_FLOW_MARK``. > > Signed-off-by: Pavan Nikhilesh > Acked-by: Andrew Rybchenko > Acked-by: Jerin Jacob > --- > v3 Changes: > - DEV_RX_OFFLOAD_RSS -> DEV_RX_

Re: [dpdk-dev] [PATCH] eal: change max hugepage sizes to 4

On Thu, Aug 8, 2019 at 11:00 AM Gagandeep Singh wrote: > > > >HI Thomas, > > > > > > DPDK currently is supporting maximum 3 hugepage, sizes whereas > > > > > > system can support more than this e.g. > > > > > > 64K, 2M, 32M and 1G. > > > > > > > > > > You can mention ARM platform here, and that th

[dpdk-dev] [PATCH] net/ipn3ke: setup MTU when HW init

set up mtu to the minimun in tx mtu, rx mtu and IPN3KE_MAC_FRAME_SIZE_MAX. Signed-off-by: Andy Pei --- drivers/net/ipn3ke/ipn3ke_ethdev.c | 93 ++ drivers/net/ipn3ke/ipn3ke_ethdev.h | 19 2 files changed, 112 insertions(+) diff --git a/drivers/net/ip

Re: [dpdk-dev] [EXT] Re: [PATCH] eal: change max hugepage sizes to 4

> -Original Message- > From: David Marchand > Sent: Thursday, August 8, 2019 1:03 PM > To: Hemant Agrawal > Cc: Thomas Monjalon ; Gagandeep Singh > ; dev ; Burakov, Anatoly > ; Olivier Matz ; > Andrew Rybchenko ; Nipun Gupta > ; Jerin Jacob Kollanukkaran ; > Gavin Hu ; Bruce Richardson >

Re: [dpdk-dev] [PATCH] eal: change max hugepage sizes to 4

> >HI Thomas, > > > > > DPDK currently is supporting maximum 3 hugepage, sizes whereas > > > > > system can support more than this e.g. > > > > > 64K, 2M, 32M and 1G. > > > > > > > > You can mention ARM platform here, and that this issue starts with > > > > kernel 5.2 (and I would try to mention th

[dpdk-dev] [patch v3] doc: announce API change in ethdev offload flags

From: Pavan Nikhilesh Add new offload flags ``DEV_RX_OFFLOAD_PTYPE``, ``DEV_RX_OFFLOAD_RSS`` and ``DEV_RX_OFFLOAD_FLOW_MARK``. Signed-off-by: Pavan Nikhilesh Acked-by: Andrew Rybchenko Acked-by: Jerin Jacob --- v3 Changes: - DEV_RX_OFFLOAD_RSS -> DEV_RX_OFFLOAD_RSS_HASH (anndrew). v2 Chan

Re: [dpdk-dev] [patch v2] doc: announce API change in ethdev offload flags

>> >> Add new offload flags ``DEV_RX_OFFLOAD_PTYPE``, >``DEV_RX_OFFLOAD_RSS`` >> and ``DEV_RX_OFFLOAD_FLOW_MARK``. >> >> Signed-off-by: Pavan Nikhilesh >> Acked-by: Andrew Rybchenko > >With DEV_RX_OFFLOAD_RSS to DEV_RX_OFFLOAD_RSS_HASH name >change as Andrew suggested. Sorry I missed it in previ

Re: [dpdk-dev] [PATCH] app/testpmd: fix latencystats deinit on signal

> -Original Message- > From: dev On Behalf Of agup...@marvell.com > Sent: Wednesday, August 7, 2019 6:42 PM > To: Wenzhuo Lu ; Jingjing Wu > ; Bernard Iremonger > Cc: dev@dpdk.org; sta...@dpdk.org; Amit Gupta > Subject: [dpdk-dev] [PATCH] app/testpmd: fix latencystats deinit on signal >

Re: [dpdk-dev] [PATCH v3 13/13] net/ipn3ke: add FPGA network side port MTU configuration

This patch set both line side and nic side MTU when Hardware init, and init to IPN3KE_MAC_FRAME_SIZE_MAX. -Original Message- From: Xu, Rosen Sent: Thursday, August 8, 2019 4:46 PM To: dev@dpdk.org Cc: Yigit, Ferruh ; Zhang, Tianfei ; Xu, Rosen ; Pei, Andy ; Lomartire, David ; Zhang, Qi

[dpdk-dev] [PATCH v3 12/13] net/ipn3ke: remove configuration for i40e port bonding

The ipn3ke board FPGA and i40e BDF scan has added in ifpga_rawdev, so it doesn't need to provide configuration for i40e port bonding. Signed-off-by: Rosen Xu --- drivers/net/ipn3ke/Makefile | 2 + drivers/net/ipn3ke/ipn3ke_ethdev.c | 289 drive

[dpdk-dev] [PATCH v3 13/13] net/ipn3ke: add FPGA network side port MTU configuration

Add FPGA network side port MTU configuration in initialization. Signed-off-by: Rosen Xu --- drivers/net/ipn3ke/ipn3ke_ethdev.c | 8 ++ drivers/net/ipn3ke/ipn3ke_ethdev.h | 55 ++ 2 files changed, 63 insertions(+) diff --git a/drivers/net/ipn3ke/ipn3ke_et

[dpdk-dev] [PATCH v3 08/13] raw/ifpga_rawdev/base: introducing sensor APIs

From: Tianfei Zhang Introducing sensor APIs to PMD driver for PAC N3000 card. Those sensor APIs: 1. opae_mgr_for_each_sensor() 2. opae_mgr_get_sensor_by_name() 3. opae_mgr_get_sensor_by_id() 4. opae_mgr_get_sensor_value_by_name() 5. opae_mgr_get_sensor_value_by_id() 6. opae_mgr_get_sensor_value(

[dpdk-dev] [PATCH v3 07/13] raw/ifpga_rawdev/base: add sensor support

From: Tianfei zhang The sensor devices are connected in MAX10 FPGA. we used the device tree to describe those sensor devices. Parse the device tree to get the sensor devices and add them into a list. Signed-off-by: Tianfei zhang --- drivers/raw/ifpga_rawdev/base/opae_intel_max10.c | 279 ++

[dpdk-dev] [PATCH v3 11/13] raw/ifpga_rawdev: add PCIe BDF devices tree scan

Add PCIe BDF devices tree scan for ipn3ke. Signed-off-by: Rosen Xu --- drivers/raw/ifpga_rawdev/ifpga_rawdev.c | 551 +++- drivers/raw/ifpga_rawdev/ifpga_rawdev.h | 16 + 2 files changed, 560 insertions(+), 7 deletions(-) diff --git a/drivers/raw/ifpga_rawdev/ifpga_

[dpdk-dev] [PATCH v3 09/13] raw/ifpga_rawdev/base: update SEU register definition

From: Tianfei zhang Update the SEU registser definition. Signed-off-by: Tianfei zhang --- drivers/raw/ifpga_rawdev/base/ifpga_defines.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/raw/ifpga_rawdev/base/ifpga_defines.h b/drivers/raw/ifpga_rawdev/base/ifpga_de

[dpdk-dev] [PATCH v3 10/13] raw/ifpga_rawdev: add SEU error handler

From: Tianfei zhang Add SEU interrupt support for FPGA. Signed-off-by: Tianfei zhang Signed-off-by: Rosen Xu --- drivers/raw/ifpga_rawdev/ifpga_rawdev.c | 246 1 file changed, 246 insertions(+) diff --git a/drivers/raw/ifpga_rawdev/ifpga_rawdev.c b/drivers/r

[dpdk-dev] [PATCH v3 05/13] raw/ifpga_rawdev/base: add device tree support

From: Tianfei zhang In PAC N3000 card, this is a BMC chip which using MAX10 FPGA to manage the board configuration, like sensors, flash controller, QSFP, powers. And this is a SPI bus connected between A10 FPGA and MAX10, we can access the MAX10 registers over this SPI bus. In BMC, there are abo

[dpdk-dev] [PATCH v3 06/13] raw/ifpga_rawdev/base: align the send buffer for SPI

From: Tianfei Zhang The length of send buffer of SPI bus should be 4bytes align. Signed-off-by: Tianfei Zhang --- .../raw/ifpga_rawdev/base/opae_spi_transaction.c | 40 +++--- 1 file changed, 36 insertions(+), 4 deletions(-) diff --git a/drivers/raw/ifpga_rawdev/base/opae_sp

[dpdk-dev] [PATCH v3 01/13] net/i40e: i40e support ipn3ke FPGA port bonding

In ipn3ke, each FPGA network side port bonding to an i40e pf, each i40e pf link status should get data from FPGA network, side port. This patch provide bonding relationship. Signed-off-by: Rosen Xu --- drivers/net/i40e/base/i40e_type.h | 3 +++ drivers/net/i40e/i40e_ethdev.c| 34 +++

[dpdk-dev] [PATCH v3 04/13] raw/ifpga_rawdev/base: add SEU error support

From: Tianfei zhang This patch exposes SEU error information to application then application could compare this information (128bit) with its own SMH file to know if this SEU is a fatal error or not. Signed-off-by: Tianfei zhang --- drivers/raw/ifpga_rawdev/base/ifpga_defines.h | 5 ++- d

[dpdk-dev] [PATCH v3 03/13] raw/ifpga_rawdev/base: clear pending bit

From: Tianfei zhang Every defined bit in FME_ERROR0 is RW1C. Other reserved bits are always 0 when readout and it will plan to be RW1C if needed in future. So it is safe just write the read back value to clear all the errors. Signed-off-by: Tianfei zhang --- drivers/raw/ifpga_rawdev/base/ifpga

[dpdk-dev] [PATCH v3 02/13] raw/ifpga_rawdev/base: add irq support

From: Tianfei zhang Add irq support for ifpga FME globle error, port error and uint unit. We implmented this feature by vfio interrupt mechanism. Signed-off-by: Tianfei zhang --- drivers/raw/ifpga_rawdev/base/ifpga_feature_dev.c | 61 +++ drivers/raw/ifpga_rawdev/base/ifpga

[dpdk-dev] [PATCH v3 00/13] Add PCIe AER disable and IRQ support for ipn3ke

This patch set adds PCIe AER disable and IRQ support for ipn3ke. Disable PCIe AER is very useful when FPGA reload. IRQ is used very widely in interrupt process. For ipn3ke is connect to CPU with PCIe switch, driver needs to scan all PCIe devices of ipn3ke, it also can get all i40e of card, so ipn3

Re: [dpdk-dev] [patch v2] doc: announce API change in ethdev offload flags

> -Original Message- > From: pbhagavat...@marvell.com > Sent: Thursday, August 8, 2019 1:48 PM > To: Jerin Jacob Kollanukkaran ; > step...@networkplumber.org; arybche...@solarflare.com; > hemant.agra...@nxp.com; tho...@monjalon.net; ferruh.yi...@intel.com; > bruce.richard...@intel.com; Nei

[dpdk-dev] [PATCH 19.11 V3 11/12] drivers/crypto/dpaa_sec: update DPAA iova table in dpaa_mem_vtop

dpaa_sec needs translations between physical and virtual addresses. V to P translation is relatively fast, as memory is managed in contiguous segments. The result of each V to P translation is used to update the DPAA iova table, which should be updated by a Mem event callback, but is not. Then the

[dpdk-dev] [PATCH 19.11 V3 09/12] virtio: get all pending rx packets with vectorized functions

From: Thibaut Collet The loop to read packets does not take all packets as the number of available packets (nb_used) is decremented in the loop. Take all available packets provides a performance improvement of 3%. Fixes: fc3d66212fed ("virtio: add vector Rx") Cc: sta...@dpdk.org Signed-off-by:

[dpdk-dev] [PATCH 19.11 V3 12/12] drivers/crypto/octeontx: enable unbinding for the OcteonTx crypto engines

Like for Ethernet ports, the OcteonTx crypto engines must first be unbound from their kernel module, then rebound to vfio-pci, before being usable in DPDK. As this capability is detected at runtime by dpdk-pmdinfo, add the info in the PMD registering directives. Then an external script can be use

[dpdk-dev] [PATCH 19.11 V3 10/12] drivers/crypto/openssl: use a local copy for the session contexts

Session contexts are used for temporary storage when processing a packet. If packets for the same session are to be processed simultaneously on multiple cores, separate contexts must be used. Note: with openssl 1.1.1 EVP_CIPHER_CTX can no longer be defined as a variable on the stack: it must be al

[dpdk-dev] [PATCH 19.11 V3 08/12] virtio: fix rx stats with vectorized functions

From: Thibaut Collet With vectorized functions, only the rx stats for number of packets is incremented. Update also the other statistics. Performance impact is about 2% Fixes: fc3d66212fed ("virtio: add vector Rx") Cc: sta...@dpdk.org Signed-off-by: Thibaut Collet --- drivers/net/virtio/virti

[dpdk-dev] [PATCH 19.11 V3 06/12] net/i40e: fix Tx descriptor status api

From: Olivier Matz The Tx descriptor status api was not behaving as expected. This API is used to inspect the content of the descriptors in the Tx ring to determine the length of the Tx queue. Since the software advances the tail pointer and the hardware advances the head pointer, the Tx queue i

[dpdk-dev] [PATCH 19.11 V3 07/12] net/i40e: set speed to undefined for default case in link update

From: Laurent Hardy During PF/VF link update, a default speed value of 100M will be set if get_link_info has failed or speed is unknown. Consequently if PF is put in no-carrier state, VFs will switch to "in carrier" state due to a link up + a link speed set to 100M (default value if no speed det

[dpdk-dev] [PATCH 19.11 V3 02/12] ethdev: fix description of tx descriptor status

From: Olivier Matz The API comment of rte_eth_tx_descriptor_status() was incorrect. The reference descriptor (when offset = 0) is not where the next packet will be sent, but where the latest packet has been enqueued. Fixes: 52f5cdd2e897 ("ethdev: add descriptor status API") Cc: sta...@dpdk.org

[dpdk-dev] [PATCH 19.11 V3 03/12] net/e1000: fix Tx descriptor status api (igb)

From: Olivier Matz The Tx descriptor status api was not behaving as expected. This API is used to inspect the content of the descriptors in the Tx ring to determine the length of the Tx queue. Since the software advances the tail pointer and the hardware advances the head pointer, the Tx queue i

[dpdk-dev] [PATCH 19.11 V3 04/12] net/e1000: fix Tx descriptor status api (em)

From: Olivier Matz The Tx descriptor status api was not behaving as expected. This API is used to inspect the content of the descriptors in the Tx ring to determine the length of the Tx queue. Since the software advances the tail pointer and the hardware advances the head pointer, the Tx queue i

[dpdk-dev] [PATCH 19.11 V3 05/12] net/ixgbe: fix Tx descriptor status api

From: Olivier Matz The Tx descriptor status api was not behaving as expected. This API is used to inspect the content of the descriptors in the Tx ring to determine the length of the Tx queue. Since the software advances the tail pointer and the hardware advances the head pointer, the Tx queue i

[dpdk-dev] [PATCH 19.11 V3 00/12] Miscellaneous fixes

This series of patches includes fixes for issues seen with 6WIND fast path, built on DPDK. The patches are in 6WIND version of DPDK, and should have been upstreamed a long time ago. V3 changes: fix double increment of stats in virtio vectorized functions V2 changes: fix checkpatch issues Guo Fe

[dpdk-dev] [PATCH 19.11 V3 01/12] net/ixgbevf: fix stats update after a PF reset

From: Guo Fengtian When PF is set down, in VF, the value of stats register is zero. So only increase stats when it's non zero. Fixes: af75078fece3 ('first public release') Cc: sta...@dpdk.org Signed-off-by: Guo Fengtian --- drivers/net/ixgbe/ixgbe_ethdev.c | 6 -- 1 file changed, 4 insert

[dpdk-dev] [patch v2] doc: announce API change in ethdev offload flags

From: Pavan Nikhilesh Add new offload flags ``DEV_RX_OFFLOAD_PTYPE``, ``DEV_RX_OFFLOAD_RSS`` and ``DEV_RX_OFFLOAD_FLOW_MARK``. Signed-off-by: Pavan Nikhilesh Acked-by: Andrew Rybchenko --- v2: Reword for clarity. doc/guides/rel_notes/deprecation.rst | 13 + 1 file changed, 13 in

[dpdk-dev] 19.11 Intel Roadmap

These are the features that we plan to submit for the 19.11 release: Intel(r) Ethernet 810 Series Network Adapter Enhancements: Support for loading of device-specific Dynamic Device Personalization (DDP) profiles to support new protocols, rte_flow/flow director/switch filter updates to support D

[dpdk-dev] [v2] net/i40e: fix vf runtime queues rss config

I40evf queue can not work properly with kernel pf driver. Eg. when configure 8 queues pair, only 4 queues can receive packets, and half packets will be lost if using 2 queues pair. This issue is caused by misconfiguration of look up table, use aq command to setup the lut to make it work properly.

Re: [dpdk-dev] [dpdk-stable] [PATCH 19.11 V2 08/12] virtio: fix rx stats with vectorized functions

On Thu, Aug 8, 2019 at 7:17 AM Tiwei Bie wrote: > > On Wed, Aug 07, 2019 at 05:09:17PM +0200, Thierry Herbelot wrote: > > From: Thibaut Collet > > > > With vectorized functions, only the rx stats for number of packets is > > incremented. > > Update also the other statistics. > > Performance impac

  1   2   >