RE: [PATCH v2 3/8] mbuf: fix Tx checksum offload examples

> From: Ferruh Yigit [mailto:ferruh.yi...@amd.com] > Sent: Monday, 15 April 2024 17.08 > > On 4/12/2024 3:44 PM, Morten Brørup wrote: > Mandate use of rte_eth_tx_prepare() in the mbuf Tx checksum > >> offload > examples. > >>> > >>> I strongly disagree with this change! >

Re: How to remove a word from the dictionary.txt we are using in the CI/CD pipeline.

We have such exception already: devtools/build-dict.sh 16/04/2024 00:19, Patrick Robb: > Hi Aditya, > > I don't run these CI checks, but I think the checkpatch script in CI > is using https://github.com/codespell-project/codespell as the > dictionary provider. So, one possibility is sugg

Re: [PATCH] mailmap: clean up vmware entries

16/04/2024 01:39, Jochen Behrens: > VMware was acquired by Broadcom, and vmware e-mail addresses will > be invalid in the future. > > Change e-mail to broadcom.com for those still in the company and > remove the other entries. > > Signed-off-by: Jochen Behrens > --- > -Bharat Mota > +Bharat Mot

[PATCH v1] crypto: fix build issues on crypto callbacks macro undefined

Crypto callbacks macro is defined with value 1 and being used with ifdef, on config value is changed to 0 to disable, crypto callback changes still being compiled. Defined crypto callbacks macro without value, undef to disable Wrapped crypto callback changes with RTE_CRYPTO_CALLBACKS macro to fix

RE: [PATCH] crypto: fix crypto callbacks macro

> -Original Message- > From: Power, Ciara > Sent: Monday, April 15, 2024 1:36 PM > To: dev@dpdk.org > Cc: Power, Ciara ; Gujjar, Abhinandan S > ; sta...@dpdk.org; Akhil Goyal > ; Fan Zhang > Subject: [PATCH] crypto: fix crypto callbacks macro > > The crypto callbacks macro is being us

Re: [PATCH 1/5] dts: fix InteractiveShell command prompt filtering

On Fri, Apr 12, 2024 at 1:11 PM Luca Vizzarro wrote: > > When sending a command using an instance of InteractiveShell the output > is meant to filter out the leading shell prompt. The filtering logic is > present but the line is appended anyways. > I don't think that's what's happening here. The

Re: [PATCH 2/5] dts: skip first line of send_command output

On Fri, Apr 12, 2024 at 1:11 PM Luca Vizzarro wrote: > > The first line of the InteractiveShell send_command method is generally > the command input field. This sometimes is unwanted, therefore this > commit enables the possibility of omitting the first line from the > returned output. > Oh, the

RE: [PATCH 2/2] dma/cnxk: remove completion pool

> -Original Message- > From: pbhagavat...@marvell.com > Sent: Saturday, April 6, 2024 3:43 PM > To: Jerin Jacob ; Vamsi Krishna Attunuru > ; Pavan Nikhilesh Bhagavatula > ; Shijith Thotton > Cc: dev@dpdk.org > Subject: [PATCH 2/2] dma/cnxk: remove completion pool > > From: Pavan Nikhi

Re: [PATCH 3/5] dts: add parsing utility module

On Fri, Apr 12, 2024 at 1:11 PM Luca Vizzarro wrote: > > Adds parsing text into a custom data structure. It provides a new > `TextParser` dataclass to be inherited. This implements the `parse` > method, which combined with the parser functions, it can automatically > parse the value for each field

Re: [PATCH 4/5] dts: add `show port info` command to TestPmdShell

On Fri, Apr 12, 2024 at 1:11 PM Luca Vizzarro wrote: > > Add a new TestPmdPort data structure to represent the output > returned by `show port info`, which is implemented as part of > TestPmdShell. > > The TestPmdPort data structure and its derived classes are modelled > based on the relevant test

Re: [PATCH 5/5] dts: add `show port stats` command to TestPmdShell

On Fri, Apr 12, 2024 at 1:11 PM Luca Vizzarro wrote: > > Add a new TestPmdPortStats data structure to represent the output > returned by `show port stats`, which is implemented as part of > TestPmdShell. > > Bugzilla ID: 1407 > > Signed-off-by: Luca Vizzarro > Reviewed-by: Paul Szczepanek Acked

RE: [PATCH v1] crypto: fix build issues on crypto callbacks macro undefined

> -Original Message- > From: Kundapura, Ganapati > Sent: Tuesday, April 16, 2024 1:42 PM > To: dev@dpdk.org > Cc: Gujjar, Abhinandan S ; Power, Ciara > ; gak...@marvell.com; fanzhang@gmail.com > Subject: [PATCH v1] crypto: fix build issues on crypto callbacks macro > undefined > >

RE: [PATCH v2 3/8] mbuf: fix Tx checksum offload examples

> > > > > > > > > > Mandate use of rte_eth_tx_prepare() in the mbuf Tx > > checksum > > > > offload > > > > > > > > > > examples. > > > > > > > > > > > > > > > > > > I strongly disagree with this change! > > > > > > > > > > > > > > > > > > It will cause a huge performance degradation for shaping

RE: [PATCH v2 3/8] mbuf: fix Tx checksum offload examples

> > From: Ferruh Yigit [mailto:ferruh.yi...@amd.com] > > Sent: Monday, 15 April 2024 17.08 > > > > On 4/12/2024 3:44 PM, Morten Brørup wrote: > > Mandate use of rte_eth_tx_prepare() in the mbuf Tx checksum > > >> offload > > examples. > > >>> > > >>> I strongly disagree

[PATCH v2] app/testpmd: fix lcore ID restriction

With modern CPUs, it is possible to have higher CPU count thus we can have higher RTE_MAX_LCORES. In testpmd application, the current config forwarding cores option "--nb-cores" is hard limited to 255. The patch fixes this constraint and also adjusts the lcore data structure to 32-bit to align wit

RE: [PATCH] common/mlx5: fix unsigned signed mismatch warning

> -Original Message- > From: Tyler Retzlaff > Sent: Tuesday, April 16, 2024 01:10 > To: dev@dpdk.org > Cc: Dariusz Sosnowski ; Dekel Peled > ; Matan Azrad ; Ori Kam > ; Suanming Mou ; Slava > Ovsiienko ; Tyler Retzlaff > ; dek...@mellanox.com; Matan Azrad > ; sta...@dpdk.org > Subject: [PA

[DPDK/ethdev Bug 1416] net/af_packet: tx_burst() can modify packets

https://bugs.dpdk.org/show_bug.cgi?id=1416 Bug ID: 1416 Summary: net/af_packet: tx_burst() can modify packets Product: DPDK Version: 24.03 Hardware: All OS: All Status: UNCONFIRMED Severity: normal

RE: [PATCH v2 3/8] mbuf: fix Tx checksum offload examples

> > > > > > > > > > > Mandate use of rte_eth_tx_prepare() in the mbuf Tx > > > checksum > > > > > offload > > > > > > > > > > > examples. > > > > > > > > > > > > > > > > > > > > I strongly disagree with this change! > > > > > > > > > > > > > > > > > > > > It will cause a huge performance degradat

Re: [PATCH 1/5] dts: fix InteractiveShell command prompt filtering

Thank you for your review Juraj! On 16/04/2024 09:40, Juraj Linkeš wrote: When sending a command using an instance of InteractiveShell the output is meant to filter out the leading shell prompt. The filtering logic is present but the line is appended anyways. I don't think that's what's happ

Re: [PATCH 2/5] dts: skip first line of send_command output

On 16/04/2024 09:48, Juraj Linkeš wrote: Oh, the first commit message was confusing. It said leading prompt which I understood to be the first prompt (the one with the command). I see that this commit actually addresses what I thought the first commit was trying to do. Yes, my bad! -def s

Re: [PATCH 3/5] dts: add parsing utility module

On 16/04/2024 09:59, Juraj Linkeš wrote: From this commit message, I don't know why we're adding the module. What are we going to use it for? Since you mentioned you'll send a v2, I'll wait with review after that as I think it'll make it a bit easier to review. Ack. Will rewrite the commit bod

Re: [PATCH 4/5] dts: add `show port info` command to TestPmdShell

On 16/04/2024 10:03, Juraj Linkeš wrote: +@dataclass +class TestPmdPort(TextParser): This and the classes above are missing docstrings. Ack. @@ -225,6 +664,38 @@ def set_forward_mode(self, mode: TestPmdForwardingModes, verify: bool = True): f"Test pmd failed to set fwd mo

[PATCH v1 1/2] test/dma: update the sg test to verify wrap around case

Run the sg test in a loop to verify wrap around case. Total number commands submitted to be more than the number descriptors allocated to verify the scenario. Signed-off-by: Vidya Sagar Velumuri diff --git a/app/test/test_dmadev.c b/app/test/test_dmadev.c index 143e1bcd68..7462e90831 100644 ---

[PATCH v1 2/2] test/dma: add functions to verify zero and one fill

Add test cases to verify zero fill and one fill Signed-off-by: Vidya Sagar Velumuri diff --git a/app/test/test.h b/app/test/test.h index 15e23d297f..0ca6519f6e 100644 --- a/app/test/test.h +++ b/app/test/test.h @@ -27,6 +27,10 @@ #include +#ifndef ARRAY_SIZE +#define ARRAY_SIZE(x) (sizeof(

[PATCH] doc/sample_app_ug: add baseline mode

Updating PMD Power Management Mode section of the l3fwd-power sample app to add baseline option as this was missing from the original commit. Fixes: a9ea60702ecb ("examples/l3fwd-power: add baseline PMD management mode") Signed-off-by: Karen Kelly --- doc/guides/sample_app_ug/l3_forward_power_m

[PATCH v2] usertools: add telemetry exporter

For now the telemetry socket is local to the machine running a DPDK application. Also, there is no official "schema" for the exposed metrics. Add a framework and a script to collect and expose these metrics to telemetry and observability agree gators such as Prometheus, Carbon or Influxdb. The expo

[PATCH 0/5] use RTE_DIM where possible

There is macro for computing the number of elements in an array RTE_DIM. But it is not used in many places where it could be. Based on similar coccinelle script in Linux. Stephen Hemminger (5): cocci: add script to use RTE_DIM app: use RTE_DIM lib: use RTE_DIM examples _use RTE_DIM drive

[PATCH 1/5] cocci: add script to use RTE_DIM

New script to find RTE_DIM should be used. Signed-off-by: Stephen Hemminger --- devtools/cocci/rte-dim.cocci | 23 +++ 1 file changed, 23 insertions(+) create mode 100644 devtools/cocci/rte-dim.cocci diff --git a/devtools/cocci/rte-dim.cocci b/devtools/cocci/rte-dim.cocci n

[PATCH 2/5] app: use RTE_DIM

Use RTE_DIM instead of computing directly with sizeof. Patch automatically generated via cocci/rte_dim.cocci. Signed-off-by: Stephen Hemminger --- app/test-flow-perf/main.c | 4 +-- app/test/test_ethdev_link.c| 3 +- app/test/test_security_inline_macsec.c | 47

[PATCH 3/5] lib: use RTE_DIM

Use RTE_DIM instead of computing directly with sizeof. Patch automatically generated via cocci/rte_dim.cocci. Signed-off-by: Stephen Hemminger --- lib/cmdline/cmdline_vt100.c | 4 +++- lib/latencystats/rte_latencystats.c | 3 +-- 2 files changed, 4 insertions(+), 3 deletions(-) diff --g

[PATCH 4/5] examples _use RTE_DIM

Use RTE_DIM instead of computing directly with sizeof. Patch automatically generated via cocci/rte_dim.cocci. Signed-off-by: Stephen Hemminger --- examples/l3fwd/main.c | 3 +-- examples/qos_sched/init.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/examples/l3fwd/m

[PATCH 5/5] drivers/net: use RTE_DIM

Use RTE_DIM instead of computing directly with sizeof. Patch automatically generated via cocci/rte_dim.cocci. Code in base/ subdirectory manually excluded. Signed-off-by: Stephen Hemminger --- drivers/net/ark/ark_pktchkr.c | 2 +- drivers/net/ark/ark_pktgen.c | 2 +- drivers/n

Re: [PATCH 2/5] app: use RTE_DIM

On Tue, Apr 16, 2024 at 08:19:28AM -0700, Stephen Hemminger wrote: > Use RTE_DIM instead of computing directly with sizeof. > Patch automatically generated via cocci/rte_dim.cocci. > > Signed-off-by: Stephen Hemminger > --- Acked-by: Tyler Retzlaff

[PATCH 0/4] net/mlx5: some unrelated fixes and improvements

This patch-set groups some unrelated fixes and improvements in MLX5 PMD code. Michael Baum (4): net/mlx5: fix secondary process port close net/mlx5/hws: fix GENEVE option class partial mask net/mlx5/hws: add fragment packet ID matching support net/mlx5/hws: remove table type DONTCARE dri

[PATCH 3/4] net/mlx5/hws: add fragment packet ID matching support

Add HWS support of IPv4 fragment packet id field matching. Signed-off-by: Michael Baum Reviewed-by: Alex Vesker Acked-by: Matan Azrad --- drivers/net/mlx5/hws/mlx5dr_definer.c | 11 ++- drivers/net/mlx5/hws/mlx5dr_definer.h | 2 ++ 2 files changed, 12 insertions(+), 1 deletion(-) dif

[PATCH 1/4] net/mlx5: fix secondary process port close

The "mlx5_dev_close()" function is used for both primary and secondary processes. If secondary process use this function after primary process is closed, the priv structure isn't valid anymore. The function is accessing priv structure to get "sh" pointer in part shared between processes causing a

[PATCH 4/4] net/mlx5/hws: remove table type DONTCARE

This patch removes the "MLX5DR_TABLE_TYPE_DONTCARE" enum value and use the correct type instead even for places the type is "don't care". Signed-off-by: Michael Baum Reviewed-by: Alex Vesker Acked-by: Dariusz Sosnowski --- drivers/net/mlx5/hws/mlx5dr.h | 1 - drivers/net/mlx5/hws/mlx5

[PATCH 2/4] net/mlx5/hws: fix GENEVE option class partial mask

When GENEVE option parser is configured, the class field has 3 optional modes: 1. ignored - ignore this field. 2. fixed - this field is part of option identifier along with type field. In this mode, the exact value is provided in "spec" field during pattern template creation a

Re: [PATCH 3/5] lib: use RTE_DIM

On Tue, Apr 16, 2024 at 08:19:29AM -0700, Stephen Hemminger wrote: > Use RTE_DIM instead of computing directly with sizeof. > Patch automatically generated via cocci/rte_dim.cocci. > > Signed-off-by: Stephen Hemminger > --- Acked-by: Tyler Retzlaff

Re: [PATCH 4/5] examples _use RTE_DIM

On Tue, Apr 16, 2024 at 08:19:30AM -0700, Stephen Hemminger wrote: > Use RTE_DIM instead of computing directly with sizeof. > Patch automatically generated via cocci/rte_dim.cocci. > > Signed-off-by: Stephen Hemminger > --- Acked-by: Tyler Retzlaff

Re: [DPDK/ethdev Bug 1416] net/af_packet: tx_burst() can modify packets

On Tue, 16 Apr 2024 10:29:53 + bugzi...@dpdk.org wrote: > static uint16_t > eth_af_packet_tx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) > { > ... > for (i = 0; i < nb_pkts; i++) { > mbuf = *bufs++; > > ... > > /* ins

Re: [PATCH v2] net/bonding: failover of LACP with mode 4 takes long time

On 6/6/2022 3:34 PM, Gaoxiang Liu wrote: > When the primary port of bond slaves with bond mode 4 linked down, > the system id of the other slave ports channged. > It may cause some switches to renegotiate, > and the process takes a few seconds. It is not acceptable for any > Telcos. > We need sub-s

RE: [DPDK/ethdev Bug 1416] net/af_packet: tx_burst() can modify packets

> > > static uint16_t > > eth_af_packet_tx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) > > { > > ... > > for (i = 0; i < nb_pkts; i++) { > > mbuf = *bufs++; > > > > ... > > > > /* insert vlan info if necessary */ > >

[DPDK/ethdev Bug 1416] net/af_packet: tx_burst() can modify packets

https://bugs.dpdk.org/show_bug.cgi?id=1416 Konstantin Ananyev (konstantin.v.anan...@yandex.ru) changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED

[PATCH] hash: cast away atomic qualification

rte_free accepts only non-cva qualified arguments so cast away RTE_ATOMIC qualification for tbl_chng_cnt and h->tbl_chng_cnt when calling rte_free. Signed-off-by: Tyler Retzlaff --- lib/hash/rte_cuckoo_hash.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/hash/rte_cu

RE: [DPDK/ethdev Bug 1416] net/af_packet: tx_burst() can modify packets

> > > static uint16_t > > > eth_af_packet_tx(void *queue, struct rte_mbuf **bufs, uint16_t > nb_pkts) > > > { > > > ... > > > for (i = 0; i < nb_pkts; i++) { > > > mbuf = *bufs++; > > > > > > ... > > > > > > /* insert vlan info if nece

Re: [PATCH 5/5] drivers/net: use RTE_DIM

On 4/16/2024 4:19 PM, Stephen Hemminger wrote: > Use RTE_DIM instead of computing directly with sizeof. > Patch automatically generated via cocci/rte_dim.cocci. > Code in base/ subdirectory manually excluded. > > Signed-off-by: Stephen Hemminger > Updated ones looks good to me, but I can see a

[PATCH] eal: fix missing type in dtor macro expansion

RTE_FINI expansion failed to specify void * type for storage of destructor function pointer resulting it defaulting to type ``int``. Update the macro to specify ``void *`` as the type so the correct size is allocated in the segment. Fixes: 64eff943ca82 ("eal: implement constructors for MSVC") Cc:

RE: [DPDK/ethdev Bug 1416] net/af_packet: tx_burst() can modify packets

> > > > static uint16_t > > > > eth_af_packet_tx(void *queue, struct rte_mbuf **bufs, uint16_t > > nb_pkts) > > > > { > > > > ... > > > > for (i = 0; i < nb_pkts; i++) { > > > > mbuf = *bufs++; > > > > > > > > ... > > > > > > > > /*

Re: [PATCH 5/5] drivers/net: use RTE_DIM

On Tue, 16 Apr 2024 17:29:25 +0100 Ferruh Yigit wrote: > On 4/16/2024 4:19 PM, Stephen Hemminger wrote: > > Use RTE_DIM instead of computing directly with sizeof. > > Patch automatically generated via cocci/rte_dim.cocci. > > Code in base/ subdirectory manually excluded. > > > > Signed-off-by: S

RE: [DPDK/ethdev Bug 1416] net/af_packet: tx_burst() can modify packets

> > > > > static uint16_t > > > > > eth_af_packet_tx(void *queue, struct rte_mbuf **bufs, uint16_t > > > nb_pkts) > > > > > { > > > > > ... > > > > > for (i = 0; i < nb_pkts; i++) { > > > > > mbuf = *bufs++; > > > > > > > > > > ... > > > > > > > > > >

RE: [PATCH] eal: fix missing type in dtor macro expansion

> From: Tyler Retzlaff [mailto:roret...@linux.microsoft.com] > Sent: Tuesday, 16 April 2024 18.33 > > RTE_FINI expansion failed to specify void * type for storage of > destructor function pointer resulting it defaulting to type ``int``. > > Update the macro to specify ``void *`` as the type so th

[PATCH v3] dts: Change hugepage runtime config to 2MB Exclusively

The previous implementation configures and allocates hugepage sizes based on a system default. This can lead to two problems: overallocation of hugepages (which may crash the remote host), and configuration of hugepage sizes that are not recommended during runtime. This new implementation allows on

RE: [PATCH v3] dts: Change hugepage runtime config to 2MB Exclusively

> From: Nicholas Pratte [mailto:npra...@iol.unh.edu] > Sent: Tuesday, 16 April 2024 20.19 > > The previous implementation configures and allocates hugepage sizes > based on a system default. This can lead to two problems: overallocation > of > hugepages (which may crash the remote host), and confi

Re: [PATCH v3] dts: Change hugepage runtime config to 2MB Exclusively

I like the changes suggested by Morten, and much such modifications accordingly. Given that the hugepage size will likely be determined by arch going forward, looking ahead, when considering patch 1360 and ongoing config changes, I decided to create a property method in the Node object that can be

Re: [DPDK/ethdev Bug 1416] net/af_packet: tx_burst() can modify packets

On Tue, 16 Apr 2024 20:11:05 +0200 Morten Brørup wrote: > > > > > > static uint16_t > > > > > > eth_af_packet_tx(void *queue, struct rte_mbuf **bufs, uint16_t > > > > nb_pkts) > > > > > > { > > > > > > ... > > > > > > for (i = 0; i < nb_pkts; i++) { > > > > > >

RE: [DPDK/ethdev Bug 1416] net/af_packet: tx_burst() can modify packets

+TO: John W. Linville, AF_PACKET maintainer > > > > > > > static uint16_t > > > > > > > eth_af_packet_tx(void *queue, struct rte_mbuf **bufs, > uint16_t > > > > > nb_pkts) > > > > > > > { > > > > > > > ... > > > > > > > for (i = 0; i < nb_pkts; i++) { > > > > > > >

[PATCH v2] mailmap: clean up vmware entries

VMware was acquired by Broadcom, and vmware e-mail addresses will be invalid in the future. Add broadcom.com addresses for those still in the company. Signed-off-by: Jochen Behrens --- .mailmap | 14 +++--- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/.mailmap b/.mailma

Reminder - DPDK Tech Board Meeting - Tomorrow, Wed. April 17, 2024 - 8am Pacific/11am Eastern/1500h UTC

You have been invited to a recurring meeting for Data Plane Development Kit (DPDK) Agenda (read-only): https://annuel.framapad.org/p/r.0c3cc4d1e011214183872a98f6b5c7db

[PATCH v2 1/2] eventdev/dma: reorganize event DMA ops

From: Pavan Nikhilesh Re-organize event DMA ops structure to allow holding source and destination pointers without the need for additional memory, the mempool allocating memory for rte_event_dma_adapter_ops can size the structure to accommodate all the needed source and destination pointers. Add

[PATCH v2 2/2] dma/cnxk: remove completion pool

From: Pavan Nikhilesh Use DMA ops to store metadata, remove use of completion pool. Signed-off-by: Pavan Nikhilesh Acked-by: Vamsi Attunuru --- drivers/dma/cnxk/cnxk_dmadev.c | 53 ++-- drivers/dma/cnxk/cnxk_dmadev.h | 24 +-- drivers/dma/cnxk/cnxk_dmadev_f

RE: [EXTERNAL] [PATCH 2/5] app: use RTE_DIM

> Use RTE_DIM instead of computing directly with sizeof. > Patch automatically generated via cocci/rte_dim.cocci. > > Signed-off-by: Stephen Hemminger > --- > app/test-flow-perf/main.c | 4 +-- > app/test/test_ethdev_link.c| 3 +- > app/test/test_security_inline_macsec

RE: [EXTERNAL] [PATCH v2 76/83] app/test: move alignment attribute on types

> Move location of __rte_aligned(a) to new conventional location. The new > placement between {struct,union} and the tag allows the desired > alignment to be imparted on the type regardless of the toolchain being > used for both C and C++. Additionally, it avoids confusion by Doxygen > when generat

RE: [EXTERNAL] [PATCH v2 01/83] examples: move alignment attribute on types

> Move location of __rte_aligned(a) to new conventional location. The new > placement between {struct,union} and the tag allows the desired > alignment to be imparted on the type regardless of the toolchain being > used for both C and C++. Additionally, it avoids confusion by Doxygen > when generat