Re: [RFC] ethdev: introduce maximum Rx buffer size

Hi Andrew, Thanks for your review. 在 2023/8/11 20:07, Andrew Rybchenko 写道: On 8/8/23 07:02, Huisong Li wrote: The Rx buffer size stands for the size hardware supported to receive packets in one mbuf. The "min_rx_bufsize" is the minimum buffer hardware supported in Rx. Actually, some engines al

Intel Downfall mitigation

Hello all, Regarding the vulnerability recently discovered on Intel AVX CPU: https://downfall.page/ Some microcode fixes are published by Intel. Should we mitigate the vulnerability in DPDK as well? In general, AVX512 is disabled by default in DPDK. Other AVX sizes are enabled if possible

Re: 21.11.5 patches review and test

Hi, Kevin RedHat QE did not find any new issues about 21.11.5 rc1 dpdk during the tests. I tested below 18 scenarios and all got PASS on RHEL9: - Guest with device assignment(PF) throughput testing(1G hugepage size): PASS - Guest with device assignment(PF) throughput testing(2M hugepage

[PATCH] doc: add NVIDIA BlueField-3 reference link

A link to the detailed guide of BlueField-3 hardware on NVIDIA website is added to the BlueField platform guide. Signed-off-by: Thomas Monjalon --- doc/guides/platform/bluefield.rst | 1 + 1 file changed, 1 insertion(+) diff --git a/doc/guides/platform/bluefield.rst b/doc/guides/platform/bluef

Re: [PATCH] devtools: list symbols by version

On Mon, Aug 14, 2023 at 11:49 AM Thomas Monjalon wrote: > > vendredi 11 août 2023, David Marchand: > > Make it easier to list experimental symbols added in a certain version. > > While at it, add a check on map symbol files content to avoid breaking > > this listing tool. > > Is there a relation b

[PATCH v2 1/2] devtools: add check on symbol maps format

Add a check on symbol maps format (fixing ethdev at the same time). This will be required by a next commit. Signed-off-by: David Marchand --- Changes since v1: - moved this check in a separate patch, - fixed ethdev map file, --- devtools/check-symbol-maps.sh | 21 + lib/ethd

[PATCH v2 2/2] devtools: list symbols by version

Make it easier to list experimental symbols added in a certain version. Example: $ ./buildtools/map-list-symbol.sh -V 18.11 lib/eal/version.map lib/eal/version.map EXPERIMENTAL rte_dev_event_callback_process lib/eal/version.map EXPERIMENTAL rte_dev_hotplug_handle_disable lib/eal/version.map EXPERI

Re: [PATCH v5 1/8] net/rnp: add skeleton

Hi, Wenbo Cao: > --- /dev/null > +++ b/doc/guides/nics/rnp.rst > @@ -0,0 +1,43 @@ > +.. SPDX-License-Identifier: BSD-3-Clause > +Copyright(c) 2023 Mucse IC Design Ltd. > + > +RNP Poll Mode driver > +== Please keep underlining the same size as the text above. > + > +T

[PATCH v1 3/3] net/hns3: report maximum buffer size

This patch reports the maximum buffer size hardware supported. Signed-off-by: Huisong Li --- drivers/net/hns3/hns3_common.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/hns3/hns3_common.c b/drivers/net/hns3/hns3_common.c index a11ea686fd..6a34df2d75 100644 --- a/drivers/net/hn

[PATCH v1 2/3] app/testpmd: add maximum Rx buffer size display

Add maximum Rx buffer size display. Signed-off-by: Huisong Li --- app/test-pmd/config.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c index 11f3a22048..e0a11a35fe 100644 --- a/app/test-pmd/config.c +++ b/app/test-pmd/config.c @@ -880,6 +880,8

[PATCH v1 0/3] introduce maximum Rx buffer size

The Rx buffer size stands for the size hardware supported to receive packets in one mbuf. The "min_rx_bufsize" is the minimum buffer hardware supported in Rx. Actually, network engines also have the maximum buffer. So this series introduce maximum Rx buffer size in struct rte_eth_dev_info and displ

[PATCH v1 1/3] ethdev: introduce maximum Rx buffer size

The Rx buffer size stands for the size hardware supported to receive packets in one mbuf. The "min_rx_bufsize" is the minimum buffer hardware supported in Rx. Actually, some engines also have the maximum buffer specification, like, hns3. For these engines, the available data size of one mbuf in Rx

NVIDIA roadmap for 23.11

Please find below NVIDIA roadmap for 23.11 release: A. rte_flow new APIs = 1. Support color-aware mode for meters in shared actions list Support meter color updates based on traffic BW. Additionally, enable use of shared (indirect) meters as part of shared action lists 1.

Re: [PATCH] dma/skeleton: fix pthread set affinity

On Mon, Aug 14, 2023 at 6:24 PM Tyler Retzlaff wrote: > > On Sat, Aug 12, 2023 at 07:27:38AM +0530, Vipin Varghese wrote: > > In skeleton dma driver, the user can pin the dma thread to desired > > cpu core. The previous commit changed the api from set-affinity to > > get-affinity leading to thread

[PATCH v4 0/8] expand list of optional libraries

This patchset contains changes to DPDK to make the build more configurable overall. The changes can be grouped into two areas: * Firstly, there are changes to make the build of the unit tests more flexible and maintainable. These 5 patches switch the unit tests from having separate hard-coded

[PATCH v4 1/8] app/test: add new macros for various test types

Rather than just registering all tests using a single generic macro, add macros which identify the test as being of a particular type. Signed-off-by: Bruce Richardson --- app/test/test.h | 9 - 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/app/test/test.h b/app/test/test.h

[PATCH v4 2/8] app/test: tag tests with the test type

Rather than having the test types called out in the meson.build file, we can use macros to identify the test type in the C file itself and then dynamically build up the tests lists at config time. Signed-off-by: Bruce Richardson --- app/test/test_acl.c | 2 +- app/test/test

[PATCH v4 3/8] app/test: make telemetry data test buildable on windows

To enable the building of the telemetry data tests file when building on windows, we need to provide a stub implementation. That way, the test file is buildable any time the library itself is built. Signed-off-by: Bruce Richardson --- app/test/test_telemetry_data.c | 14 -- 1 file ch

[PATCH v4 4/8] app/test: build using per-file dependency matrix

Rather than using if-else constructs to selectively add or remove files from the UT build, switch to a table-based approach where each file lists out what libs or drivers it depends upon. Initial version of this table was generated via analysis of the header files included in each C file. The basi

[PATCH v4 5/8] app/test: define unit tests suites based on test macros

Rather than having the test suites listed out in the meson.build files and having to have them enabled/disabled selectively based on what libs are being built, pull the tests to run from the source files which were added to the build. Most test suites require no additional info other than the list

[PATCH v4 6/8] examples/l3fwd: make eventdev an optional dependency

With l3fwd being a very commonly used example app, and built as part of our CI tests, we need to ensure it's buildable with just about all supported DPDK configurations. To enable l3fwd application to be built when the eventdev library is disabled, we need to compile in the eventdev support condit

[PATCH v4 7/8] build: make most device classes optional

Apart from ethdev and cryptodev, which have lots of components and tests which depend on them, we can make the device class libraries optional without too much work. This patch marks: * bbdev, * compressdev, * dmadev, * eventdev, * mldev, * rawdev, * regexdev optional, and ensures that DPDK - incl

[PATCH v4 8/8] build: expand list of optional libraries

With the unit test build now with individual per-file dependencies, we can more easily expand the list of optional libraries. Add 8 new libraries to the optional list. Signed-off-by: Bruce Richardson --- lib/meson.build | 8 1 file changed, 8 insertions(+) diff --git a/lib/meson.build

Re: [PATCH v4 0/8] expand list of optional libraries

On Tue, Aug 15, 2023 at 02:13:12PM +0100, Bruce Richardson wrote: > This patchset contains changes to DPDK to make the build more configurable > overall. The changes can be grouped into two areas: > > * Firstly, there are changes to make the build of the unit tests more > flexible and maintainab

[Bug 1274] [20.11 LTS] MPRQ parameters are incorrectly adjusted and not applied properly

https://bugs.dpdk.org/show_bug.cgi?id=1274 Bug ID: 1274 Summary: [20.11 LTS] MPRQ parameters are incorrectly adjusted and not applied properly Product: DPDK Version: 20.11 Hardware: All OS: All S

Re: [PATCH v4 4/4] build: enable MSVC specific compiler options

Hello Tyler, On Fri, Aug 11, 2023 at 8:24 PM Tyler Retzlaff wrote: > > * Enable optional use of C11 atomics support. > * Enable use of C23 typeof operator. > * Explicitly force intrinsics when building with MSVC. > * Disable MSVC C runtime checks. > > Signed-off-by: Tyler Retzlaff > --- > confi

RE: 20.11.9 patches review and test

> -Original Message- > From: Luca Boccassi > Sent: Monday, August 14, 2023 2:55 PM > To: Ali Alnubani > Cc: sta...@dpdk.org; dev@dpdk.org; John McNamara > ; Raslan Darawsheh ; > NBU-Contact-Thomas Monjalon (EXTERNAL) ; Wael > Abualrub > Subject: Re: 20.11.9 patches review and test > > O

[PATCH v4 0/8] support reinit flow

From: Sinan Kaya We want to be able to call rte_eal_init() and rte_eal_cleanup() APIs back to back for maintanance reasons. Here is a summary of the code we have seen so far: 1. some code support getting called multiple times by keeping a static variable. 2. some code initializes once but never

[PATCH v4 1/8] eal: cleanup plugins data

From: Stephen Hemminger When rte_eal_cleanup is called walk through the list of shared objects loaded, and close them and free the data structure. Signed-off-by: Stephen Hemminger Signed-off-by: Sinan Kaya --- lib/eal/common/eal_common_options.c | 12 lib/eal/common/eal_options.h

[PATCH v4 2/8] eal: fixes for re-initialization issues

From: Graham Whyte reinitialize the solib link list and clean the globals holding state for parsing. Signed-off-by: Sinan Kaya Signed-off-by: Graham Whyte --- lib/eal/common/eal_common_options.c | 8 lib/eal/linux/eal.c | 4 +++- 2 files changed, 11 insertions(+), 1 d

[PATCH v4 3/8] tailq: skip init if already initialized

From: Sinan Kaya Allows tailq to be reinitialied multiple times by looking up previously registered tailqs Signed-off-by: Sinan Kaya --- lib/eal/common/eal_common_tailqs.c | 20 +--- 1 file changed, 13 insertions(+), 7 deletions(-) diff --git a/lib/eal/common/eal_common_tailqs

[PATCH v4 4/8] eal_memzone: bail out on initialized

From: Sinan Kaya Initialize memzone once and bail out if someone calls init multiple times. Signed-off-by: Sinan Kaya --- lib/eal/common/eal_common_memzone.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/lib/eal/common/eal_common_memzone.c b/lib/eal/common/eal_common_memzone.c in

[PATCH v4 5/8] memseg: init once

From: Sinan Kaya Initialize memory segments just once and bail out if called multiple times. Signed-off-by: Sinan Kaya --- lib/eal/linux/eal_memory.c | 12 +++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/lib/eal/linux/eal_memory.c b/lib/eal/linux/eal_memory.c index 9

[PATCH v4 6/8] eal_memory: skip initialization

From: Sinan Kaya Initialize heap area just once. Signed-off-by: Sinan Kaya --- lib/eal/common/eal_common_memory.c | 5 + lib/eal/common/malloc_heap.c | 7 +++ 2 files changed, 12 insertions(+) diff --git a/lib/eal/common/eal_common_memory.c b/lib/eal/common/eal_common_memory.c

[PATCH v4 7/8] eal_interrupts: don't reinitialize threads

From: Sinan Kaya Initialize interrupt thread once and keep a flag for further init. Signed-off-by: Sinan Kaya --- lib/eal/linux/eal_interrupts.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/lib/eal/linux/eal_interrupts.c b/lib/eal/linux/eal_interrupts.c index c9881143be..6a35b4ae

[PATCH v4 8/8] eal: initialize worker threads once

From: Sinan Kaya Initialize worker threads once and keep a flag for other init calls. Signed-off-by: Sinan Kaya --- lib/eal/linux/eal.c | 60 - 1 file changed, 32 insertions(+), 28 deletions(-) diff --git a/lib/eal/linux/eal.c b/lib/eal/linux/eal.c

Re: 20.11.9 patches review and test

On Tue, 15 Aug 2023 at 14:35, Ali Alnubani wrote: > > > -Original Message- > > From: Luca Boccassi > > Sent: Monday, August 14, 2023 2:55 PM > > To: Ali Alnubani > > Cc: sta...@dpdk.org; dev@dpdk.org; John McNamara > > ; Raslan Darawsheh ; > > NBU-Contact-Thomas Monjalon (EXTERNAL) ; Wae

[PATCH] app/dma-perf: fix physical address seg-fault

do_cpu_mem_copy uses DPDK API rte_mbuf_data_iova to return the start of the virtual address for both src and dst. But in case of iova mode set as PA, this results in seg-fault. This is because rte_memcpy VA address and not PA. this fix checks the mode and invoked rte_memcpy accrodingly. Bugzilla

[PATCH] app/dma-perf: fix physical address seg-fault

do_cpu_mem_copy uses DPDK API rte_mbuf_data_iova to return the start of the virtual address for both src and dst. But in case of iova mode set as PA, this results in seg-fault. This is because rte_memcpy VA address and not PA. this fix checks the iova mode and invokes rte_memcpy with the right arg

[PATCH v5 10/10] build: expand list of optional libraries

With the unit test build now with individual per-file dependencies, we can more easily expand the list of optional libraries. Add 8 new libraries to the optional list. Signed-off-by: Bruce Richardson Acked-by: Morten Brørup --- lib/meson.build | 8 1 file changed, 8 insertions(+) diff

RE: [PATCH] app/dma-perf: fix physical address seg-fault

[AMD Official Use Only - General] Apologies, marking this as `NA`. After recheck of this logic without use of ` rte_mbuf_data_iova` will result in mbuf meta-data corruption. Need to fix this in a different way. > -Original Message- > From: Vipin Varghese > Sent: Tuesday, August 15, 202

[PATCH v5 01/10] app/test: add new macros for various test types

Rather than just registering all tests using a single generic macro, add macros which identify the test as being of a particular type. Signed-off-by: Bruce Richardson Acked-by: Morten Brørup --- app/test/test.h | 9 - 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/app/test

[PATCH v5 00/10] expand list of optional libraries

This patchset contains changes to DPDK to make the build more configurable overall. The changes can be grouped into two areas: * Firstly, there are changes to make the build of the unit tests more flexible and maintainable. These 7 patches switch the unit tests from having separate hard-coded

[PATCH v5 02/10] app/test: tag tests with the test type

Rather than having the test types called out in the meson.build file, we can use macros to identify the test type in the C file itself and then dynamically build up the tests lists at config time. Signed-off-by: Bruce Richardson Acked-by: Morten Brørup --- app/test/test_acl.c

[PATCH v5 03/10] app/test: make telemetry data test buildable on windows

To enable the building of the telemetry data tests file when building on windows, we need to provide a stub implementation. That way, the test file is buildable any time the library itself is built. Signed-off-by: Bruce Richardson Acked-by: Morten Brørup --- app/test/test_telemetry_data.c | 14

[PATCH v5 04/10] app/test: build using per-file dependency matrix

Rather than using if-else constructs to selectively add or remove files from the UT build, switch to a table-based approach where each file lists out what libs or drivers it depends upon. Initial version of this table was generated via analysis of the header files included in each C file. The basi

[PATCH v5 06/10] app/test: add test case for scripted telemetry commands

The test script for calling all telemetry commands is not discoverable by checking the C files for the build. Therefore we need to add it in as a special-case test in the fast-tests group. Signed-off-by: Bruce Richardson --- app/test/suites/meson.build | 30 +

[PATCH v5 07/10] app/test: add debug test suite

Add in for all platforms, the suite of tests to dump out the structure information available. Since the commands are defined in commands.c on all OS's, do not limit their presence to just the non-windows OS's. Signed-off-by: Bruce Richardson --- app/test/suites/meson.build | 20 +

[PATCH v5 08/10] examples/l3fwd: make eventdev an optional dependency

With l3fwd being a very commonly used example app, and built as part of our CI tests, we need to ensure it's buildable with just about all supported DPDK configurations. To enable l3fwd application to be built when the eventdev library is disabled, we need to compile in the eventdev support condit

[PATCH v5 05/10] app/test: define unit tests suites based on test macros

Rather than having the test suites listed out in the meson.build files and having to have them enabled/disabled selectively based on what libs are being built, pull the tests to run from the source files which were added to the build. Most test suites require no additional info other than the list

[PATCH v5 09/10] build: make most device classes optional

Apart from ethdev and cryptodev, which have lots of components and tests which depend on them, we can make the device class libraries optional without too much work. This patch marks: * bbdev, * compressdev, * dmadev, * eventdev, * mldev, * rawdev, * regexdev optional, and ensures that DPDK - incl

[PATCH v3 1/2] devtools: add check on symbol maps format

Add a check on symbol maps format (fixing ethdev at the same time). This will be required by a next commit. Signed-off-by: David Marchand --- Changes since v2: - fixed FreeBSD build by replacing (|pattern) with (pattern)?, Changes since v1: - moved this check in a separate patch, - fixed ethdev

[PATCH v3 2/2] devtools: list symbols by version

Make it easier to list experimental symbols added in a certain version. Example: $ ./buildtools/map-list-symbol.sh -V 18.11 lib/eal/version.map lib/eal/version.map EXPERIMENTAL rte_dev_event_callback_process lib/eal/version.map EXPERIMENTAL rte_dev_hotplug_handle_disable lib/eal/version.map EXPERI

Re: [PATCH v4 0/8] support reinit flow

On Tue, 15 Aug 2023 10:50:15 -0400 ok...@kernel.org wrote: > From: Sinan Kaya > > We want to be able to call rte_eal_init() and rte_eal_cleanup() > APIs back to back for maintanance reasons. > > Here is a summary of the code we have seen so far: > > 1. some code support getting called multiple

Re: [PATCH v4 8/8] eal: initialize worker threads once

On Tue, 15 Aug 2023 10:50:23 -0400 ok...@kernel.org wrote: > From: Sinan Kaya > > Initialize worker threads once and keep a flag > for other init calls. > > Signed-off-by: Sinan Kaya > --- > lib/eal/linux/eal.c | 60 - > 1 file changed, 32 insertion

Re: [PATCH v4 2/8] eal: fixes for re-initialization issues

On Tue, 15 Aug 2023 10:50:17 -0400 ok...@kernel.org wrote: > +static uint32_t run_once; > + > int > eal_clean_runtime_dir(void) > { > @@ -505,6 +507,7 @@ eal_parse_socket_arg(char *strval, volatile uint64_t > *socket_arg) > socket_arg[i] = val; > } > > + __atomic_stor

Re: [PATCH v4 2/8] eal: fixes for re-initialization issues

On Tue, 2023-08-15 at 10:49 -0700, Stephen Hemminger wrote: > > > > + __atomic_store_n(&run_once, 0, __ATOMIC_RELAXED); > >return 0; > > > Interesting, other flags don't use atomic. Why here? > > > > And is already set elsewhere? Looking at the history, this variable used to be

Re: [PATCH v4 8/8] eal: initialize worker threads once

On Tue, 2023-08-15 at 10:46 -0700, Stephen Hemminger wrote: > > static uint32_t run_once; > > +static bool worker_initialized; > > > > > I see a pattern here. Many places are using static to > > only initialize once. Could you name these variables the > > same; suggest using run_once every

Re: [PATCH v4 0/8] support reinit flow

On Tue, 2023-08-15 at 10:45 -0700, Stephen Hemminger wrote: > > Why? > > It is not always ideal to reinitialize a DPDK process. Memory needs > > to be reinitialized, hugetables need to warm up etc. > > > > > I am familiar with the backstory of why this is desirable in your > case. > > But othe

Re: [PATCH v5 04/10] app/test: build using per-file dependency matrix

Adam from our team just raised something important about this patch and UNH CI which I missed during the RFC discussion. Presently, eal_flags_file_prefix_autotest fails on arm tx2 systems (arm people are aware of this, they think it might be a memory leak from mlx5 driver), so at their request we a

RE: [EXT] [PATCH v1 0/3] Add support for inter-domain DMA operations

Hi Anatoly > -Original Message- > From: Anatoly Burakov > Sent: Friday, August 11, 2023 9:15 AM > To: dev@dpdk.org > Cc: bruce.richard...@intel.com > Subject: [EXT] [PATCH v1 0/3] Add support for inter-domain DMA > operations > > External Email > > --

Re: [PATCH v4 0/8] support reinit flow

On Tue, 15 Aug 2023 14:58:03 -0400 Sinan Kaya wrote: > > > > As a compromise, can this restart be officially tagged as > > unsupported. > > any pointers how to do this? > > I have no idea how to mark something unsupported in code. > If this is acceptable in cover letter, I'm happy to do that

RE: [PATCH v2] net/ice: write rx timestamp to the first mbuf segment in scattered rx

> -Original Message- > From: Martin Weiser > Sent: Tuesday, August 8, 2023 7:40 PM > To: dev@dpdk.org > Cc: Martin Weiser ; Yang, Qiming > ; Zhang, Qi Z > Subject: [PATCH v2] net/ice: write rx timestamp to the first mbuf segment in > scattered rx > > Previously, the rx timestamp was w

RE: [PATCH] bus/pci: fix legacy device IO port map in secondary process

Hi, > -Original Message- > From: Gupta, Nipun > Sent: 2023年8月13日 14:15 > To: Ma, WenwuX ; dev@dpdk.org > Cc: david.march...@redhat.com; maxime.coque...@redhat.com; Xia, > Chenbo ; Li, Miao ; Ling, WeiX > ; sta...@dpdk.org > Subject: Re: [PATCH] bus/pci: fix legacy device IO port map in se

RE: [PATCH] app/dma-perf: fix physical address seg-fault

Hi Vipin, Would using 'rte_pktmbuf_mtod' in 'do_cpu_mem_copy' function resolve the issue? Thanks, Anoob > -Original Message- > From: Varghese, Vipin > Sent: Tuesday, August 15, 2023 8:45 PM > To: Varghese, Vipin ; tho...@monjalon.net; > dev@dpdk.org > Cc: Yigit, Ferruh ; cheng1.ji...@in

[PATCH v5 0/2] replace master/slave to main/member

This patch series aims to replace all the master/slave to main/member in the bonding PMD. --- v5: * Rebase to DPDK v23.11. v4: * Fix the doc compile problem. v3: * Fix the check warning about 'CamelCase'. v2: * Modify related doc. * Add 'RTE_DEPRECATED' to related APIs. --- Long Wu (2): ethdev

[PATCH v5 1/2] ethdev: remove bonding deprecated macro

From: Long Wu The macro RTE_ETH_DEV_BONDED_SLAVE was marked as deprecated in v23.07, we can remove it. Signed-off-by: Long Wu Reviewed-by: Chaoyong He --- doc/guides/rel_notes/deprecation.rst | 5 + doc/guides/rel_notes/release_23_11.rst | 2 ++ lib/ethdev/rte_ethdev.h|

[PATCH v2 00/12] net/cpfl: support port representor

From: Beilei Xing 1. code refine for representor support 2. support port representor v2 changes: - Remove representor data path. - Fix coding style. Beilei Xing (12): net/cpfl: refine devargs parse and process net/cpfl: introduce interface structure net/cpfl: add cp channel net/cpfl:

[PATCH v2 01/12] net/cpfl: refine devargs parse and process

From: Beilei Xing 1. Keep devargs in adapter. 2. Refine handling the case with no vport be specified in devargs. 3. Separate devargs parse and devargs process Signed-off-by: Qi Zhang Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_ethdev.c | 154 ++--- drivers

[PATCH v2 03/12] net/cpfl: add cp channel

From: Beilei Xing Add cpchnl header file. Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_cpchnl.h | 321 + 1 file changed, 321 insertions(+) create mode 100644 drivers/net/cpfl/cpfl_cpchnl.h diff --git a/drivers/net/cpfl/cpfl_cpchnl.h b/drivers/net/cpfl/

[PATCH v2 02/12] net/cpfl: introduce interface structure

From: Beilei Xing Introduce cplf interface structure to distinguish vport and port representor. Signed-off-by: Qi Zhang Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_ethdev.c | 3 +++ drivers/net/cpfl/cpfl_ethdev.h | 16 2 files changed, 19 insertions(+) diff --git a

[PATCH v2 04/12] net/cpfl: enable vport mapping

From: Beilei Xing 1. Handle cpchnl event for vport create/destroy 2. Use hash table to store vport_id to vport_info mapping 3. Use spinlock for thread safe. Signed-off-by: Qi Zhang Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_ethdev.c | 157 + drivers/n

[PATCH v2 05/12] net/cpfl: parse representor devargs

From: Beilei Xing Format: [[c]pf]vf controller_id: 0 : xeon (default) 1: acc pf_id: 0 : apf (default) 1 : cpf Example: representor=c0pf0vf[0-3] -- xeon > apf > vf 0,1,2,3 same as pf0vf[0-3] and vf[0-3] if omit default value. representor=c0pf0 -- xeon> apf same a

[PATCH v2 06/12] net/cpfl: support probe again

From: Beilei Xing Only representor will be parsed for probe again. Signed-off-by: Qi Zhang Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_ethdev.c | 69 +++--- 1 file changed, 56 insertions(+), 13 deletions(-) diff --git a/drivers/net/cpfl/cpfl_ethdev.c b/dr

[PATCH v2 07/12] net/cpfl: create port representor

From: Beilei Xing Track representor request in a whitelist. Representor will only be created for active vport. Signed-off-by: Jingjing Wu Signed-off-by: Qi Zhang Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_ethdev.c | 107 --- drivers/net/cpfl/cpfl_ethdev.h | 34 +++ d

[PATCH v2 08/12] net/cpfl: support vport list/info get

From: Beilei Xing Support cp channel ops CPCHNL2_OP_CPF_GET_VPORT_LIST and CPCHNL2_OP_CPF_GET_VPORT_INFO. Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_ethdev.h | 8 drivers/net/cpfl/cpfl_vchnl.c | 72 ++ drivers/net/cpfl/meson.build | 1 + 3 fi

[PATCH v2 09/12] net/cpfl: update vport info before creating representor

From: Beilei Xing Get port representor's vport list and update vport_map_hash before creating the port representor. Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_ethdev.c | 2 +- drivers/net/cpfl/cpfl_ethdev.h | 3 + drivers/net/cpfl/cpfl_representor.c | 124 ++

[PATCH v2 10/12] net/cpfl: refine handle virtual channel message

From: Beilei Xing Refine handle virtual channel event message. Signed-off-by: Qi Zhang Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_ethdev.c | 46 -- 1 file changed, 22 insertions(+), 24 deletions(-) diff --git a/drivers/net/cpfl/cpfl_ethdev.c b/driver

[PATCH v2 11/12] net/cpfl: support link update for representor

From: Beilei Xing Add link update ops for representor. Signed-off-by: Jingjing Wu Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_ethdev.h | 1 + drivers/net/cpfl/cpfl_representor.c | 21 + 2 files changed, 22 insertions(+) diff --git a/drivers/net/cpfl/cpfl_et

[PATCH v2 12/12] net/cpfl: support Rx/Tx queue setup for representor

From: Beilei Xing Add dummy Rx/Tx queue setup functions for representor. Signed-off-by: Jingjing Wu Signed-off-by: Beilei Xing --- drivers/net/cpfl/cpfl_representor.c | 26 ++ 1 file changed, 26 insertions(+) diff --git a/drivers/net/cpfl/cpfl_representor.c b/drivers