Re: [PATCH 2/3] power: defer lcore variable allocation

On 2024-12-05 18:57, David Marchand wrote: The lcore variable in this code unit is only used through rte_power_ethdev_pmgmt_queue_*() public symbols. Defer the unconditional lcore variable allocation in those symbols. Fixes: 130643319579 ("power: keep per-lcore state in lcore variable") Cc: sta

Re: [PATCH 3/3] eal/x86: defer power intrinsics variable allocation

On 2024-12-05 18:57, David Marchand wrote: The lcore variable in this code unit is only used through rte_power_monitor*() public symbols. Defer the unconditional lcore variable allocation in those symbols. Fixes: 18b5049ab4fe ("eal/x86: keep power intrinsics state in lcore variable") Cc: sta...

RE: [PATCH v3 2/7] drivers/bus: eliminate dependency on non-portable __SIZEOF_LONG__

> > Macro __SIZEOF_LONG__ is not standardized and MSVC does not define it. > Therefore the errors below are seen with MSVC: > > ../lib/mldev/mldev_utils_scalar.c(465): error C2065: > '__SIZEOF_LONG__': undeclared identifier > ../lib/mldev/mldev_utils_scalar.c(478): error C2051: > case

[PATCH v3 3/3] net/macb: fix logic error in macb_rxq_rearm function

Fixed an issue in the macb_rxq_rearm function where descriptors were incorrectly set when mbuf allocation failed, leading to abnormal behavior of the network card. Fixes: 1e3e9ebd1e37 ("net/macb: add NEON vectorized Rx/Tx") Cc: liwench...@phytium.com.cn Signed-off-by: Wencheng Li --- drivers/ne

[PATCH] net/sxe: add a basic directory structure and doc

Add a basic directory structure and documentation for a new nic polling mode driver. Signed-off-by: Jie Liu --- MAINTAINERS | 6 app/test-pmd/meson.build| 3 ++ doc/guides/nics/features/sxe.ini| 10 ++ doc/guides/nics/features/sxe_vf

Re: GCP cloud : Virtio-PMD performance Issue

Hi Mukul, On 12/5/24 23:54, Mukul Sinha wrote: Thanks @maxime.coque...@redhat.com Have included dev@dpdk.org On Fri, Dec 6, 2024 at 2:11 AM Maxime Coquelin mailto:maxime.coque...@redhat.com>> wrote: Hi Mukul, DPDK upstream m

Re: [PATCH 1/3] random: defer seeding to EAL init

On 2024-12-05 18:57, David Marchand wrote: The RNG is documented as being seeded as part of EAL init. /** * Seed the pseudo-random generator. * * The generator is automatically seeded by the EAL init with a timer * value. It may need to be re-seeded by the user with a real random * val

Re: [PATCH 0/3] Defer lcore variables allocation

On 2024-12-05 18:57, David Marchand wrote: As I had reported in rc2, the lcore variables allocation have a noticeable impact on applications consuming DPDK, even when such applications does not use DPDK, or use features associated to some lcore variables. While the amount has been reduced in a r

[PATCH v3 2/3] net/macb: add NEON vectorized Rx/Tx

To optimize Rx/Tx burst process, add NEON vector instructions on arm architecture. Signed-off-by: Wencheng Li --- drivers/net/macb/macb_rxtx.c | 2 + drivers/net/macb/macb_rxtx_vec_neon.c | 672 ++ drivers/net/macb/meson.build | 4 + 3 files

[PATCH v2] net/mlx5: fix hypervisor detection in VLAN workaround

The mlx5 PMD provides a specific workaround for the VMware ESXi hypervisor, enabling on-demand routing configuration to virtual machines. This workaround activates when the device type is a Virtual Function and either an ESXi hypervisor is detected or the hypervisor type is unknown. For non-x86 ar

RE: [PATCH v3 2/7] drivers/bus: eliminate dependency on non-portable __SIZEOF_LONG__

> > > Macro __SIZEOF_LONG__ is not standardized and MSVC does not define it. > > > Therefore the errors below are seen with MSVC: > > > > > > ../lib/mldev/mldev_utils_scalar.c(465): error C2065: > > > '__SIZEOF_LONG__': undeclared identifier > > > ../lib/mldev/mldev_utils_scalar.c(478): error

RE: [PATCH v3 2/7] drivers/bus: eliminate dependency on non-portable __SIZEOF_LONG__

> -Original Message- > From: Konstantin Ananyev > Sent: Friday, December 6, 2024 4:41 PM > To: Andre Muezerie > Cc: dev@dpdk.org > Subject: RE: [PATCH v3 2/7] drivers/bus: eliminate dependency on non-portable > __SIZEOF_LONG__ > > > > > > Macro __SIZEOF_LONG__ is not standardized an

[PATCH 1/2] lib/dmadev: eliminate undefined behavior

MSVC compiler issues warnings like the one below: lib\dmadev\rte_dmadev_trace_fp.h(36): warning C5101: use of preprocessor directive in function-like macro argument list is undefined behavior Indeed, looking at C99 section 6.10.3 Macro replacement, paragraph 11: "If there are sequences of

[PATCH 2/2] lib/dmadev: enable dmadev lib to be compiled with MSVC

With the outstanding issues preventing dmadev from being compiled with MSVC being fixed by this series, the lib can be enabled to be compiled with MSVC. Signed-off-by: Andre Muezerie --- lib/dmadev/meson.build | 6 -- 1 file changed, 6 deletions(-) diff --git a/lib/dmadev/meson.build b/lib/

[PATCH 0/2] enable dmadev lib to be compiled with MSVC

The first patch eliminates undefined behavior in the dmadev lib: lib\dmadev\rte_dmadev_trace_fp.h(36): warning C5101: use of preprocessor directive in function-like macro argument list is undefined behavior The second patch enables the dmadev library to be compiled with MSVC. Andre Mueze

Scheduled maintenance for git.dpdk.org

Hello, This is to inform you about upcoming scheduled maintenance for one of the dpdk.org servers. The maintenance will start on Sunday at approximately 8:00 AM UTC and may last up to 2 hours. During the maintenance window, you may experience disruptions while accessing the following sites/ser

Re: [PATCH v3 2/7] drivers/bus: eliminate dependency on non-portable __SIZEOF_LONG__

On Fri, Dec 06, 2024 at 04:41:16PM +, Konstantin Ananyev wrote: > > > > > Macro __SIZEOF_LONG__ is not standardized and MSVC does not define it. > > > > Therefore the errors below are seen with MSVC: > > > > > > > > ../lib/mldev/mldev_utils_scalar.c(465): error C2065: > > > > '__SIZEOF_LON

Re: [PATCH v3 2/7] drivers/bus: eliminate dependency on non-portable __SIZEOF_LONG__

On Fri, Dec 06, 2024 at 12:22:42PM +, Konstantin Ananyev wrote: > > > > > > Macro __SIZEOF_LONG__ is not standardized and MSVC does not define it. > > Therefore the errors below are seen with MSVC: > > > > ../lib/mldev/mldev_utils_scalar.c(465): error C2065: > > '__SIZEOF_LONG__': undec

RE: [PATCH v16 1/4] lib: add generic support for reading PMU events

> > Add support for programming PMU counters and reading their values > in runtime bypassing kernel completely. > > This is especially useful in cases where CPU cores are isolated > i.e run dedicated tasks. In such cases one cannot use standard > perf utility without sacrificing latency and pe

Re: [PATCH 0/3] Defer lcore variables allocation

06/12/2024 12:01, Mattias Rönnblom: > On 2024-12-05 18:57, David Marchand wrote: > In retrospect, maybe the offset between lcore variable instances could > have been encoded into the handle, and thus one could use > different-sized offset for different variables. Yes it would allow to allocate a

[PATCH v11 6/7] app/test: add unit tests for soring API

Add both functional and stess test-cases for soring API. Stress test serves as both functional and performance test of soring enqueue/dequeue/acquire/release operations under high contention (for both over committed and non-over committed scenarios). Signed-off-by: Eimear Morrissey Signed-off-by:

[PATCH v11 4/7] ring: make dump function more verbose

From: Eimear Morrissey The current rte_ring_dump function uses the generic rte_ring_headtail structure to access head/tail positions. This is incorrect for the RTS case where the head is stored in a different offset in the union of structs. Switching to a separate function for each sync type allo

[PATCH v11 5/7] ring/soring: introduce Staged Ordered Ring

Staged-Ordered-Ring (SORING) provides a SW abstraction for 'ordered' queues with multiple processing 'stages'. It is based on conventional DPDK rte_ring, re-uses many of its concepts, and even substantial part of its code. It can be viewed as an 'extension' of rte_ring functionality. In particular,

[PATCH v11 2/7] ring: common functions for 'move head' ops

Note upfront: that change doesn't introduce any functional or performance changes. It is just a code-reordering for: - code deduplication - ability in future to re-use the same code to introduce new functionality For each sync mode corresponding move_prod_head() and move_cons_head() are nearly i

[PATCH v11 3/7] ring: make copying functions generic

Note upfront: that change doesn't introduce any functional or performance changes. It is just a code-reordering for: - improve code modularity and re-usability - ability in future to re-use the same code to introduce new functionality There is no real need for enqueue_elems()/dequeue_elems() to

[PATCH v11 0/7] Stage-Ordered API and other extensions for ring library

Testing coverage (passed): x86_64, i686, PPC, ARM Would like to express my gratitude to all community members who helped with testing it on different platforms, in particular: David Christensen Cody Cheng Patrick Robb Phanendra Vukkisala Chengwen Feng v10 - v11 - rebased on top of 25.03-r

[PATCH v11 1/7] test/ring: fix failure with custom number of lcores

ring_stress_autotest fails to initialize the ring when RTE_MAX_LCORE value is not a number of 2. There is a flaw in calculation required number of elements in the ring. Fix it by aligning number of elements to next power of 2. Fixes: bf28df24e915 ("test/ring: add contention stress test") Cc: sta..

[PATCH v11 7/7] test: add stress test suite

Add a new test suite which purpose is to run 'stress' tests: main purpose is put a pressure to dpdk sync algorithms to flag their misbehaving/slowdown/etc. Right now it consists from just 2 test-cases: meson test --suite stress-tests --list DPDK:stress-tests / ring_stress_autotest DPDK:stress-tests