DPDK patches and discussions
 help / color / mirror / Atom feed
* NVIDIA roadmap for 22.03
@ 2021-12-16  7:22 Lior Margalit
  2021-12-16 16:45 ` Lior Margalit
  0 siblings, 1 reply; 2+ messages in thread
From: Lior Margalit @ 2021-12-16  7:22 UTC (permalink / raw)
  To: dev

[-- Attachment #1: Type: text/plain, Size: 923 bytes --]

Please find below NVIDIA roadmap for 22.03 release:



rte_flow new APIs:

================

[1] Support for template API (item and action) to optimize the insertion rate.

The combination of item and action templates along with shared flow rule attributes allows the PMD/HW to prepare all the resources needed for efficient flow rules creation in the datapath.

[2] Support for asynchronous queue-based flow rules management mechanism for applications.

A flow rule creation/destruction is enqueued on the specified flow queue and offloaded asynchronously to the hardware.



mlx5 PMD updates:

==================

[3] Support the rte_flow changes listed above.



rte_mempool updates:

===================

[4] Optimize the hugepage initialization to speed up the application start-up time.

testpmd updates:

================

[5] Support the changes in rte_flow listed above.




[-- Attachment #2: Type: text/html, Size: 3520 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

* RE: NVIDIA roadmap for 22.03
  2021-12-16  7:22 NVIDIA roadmap for 22.03 Lior Margalit
@ 2021-12-16 16:45 ` Lior Margalit
  0 siblings, 0 replies; 2+ messages in thread
From: Lior Margalit @ 2021-12-16 16:45 UTC (permalink / raw)
  To: dev

[-- Attachment #1: Type: text/plain, Size: 1260 bytes --]

Added one item below for mlx5 PMD.

From: Lior Margalit
Sent: Thursday, December 16, 2021 9:22 AM
To: dev@dpdk.org
Subject: NVIDIA roadmap for 22.03


Please find below NVIDIA roadmap for 22.03 release:



rte_flow new APIs:

================

[1] Support for template API (item and action) to optimize the insertion rate.

The combination of item and action templates along with shared flow rule attributes allows the PMD/HW to prepare all the resources needed for efficient flow rules creation in the datapath.

[2] Support for asynchronous queue-based flow rules management mechanism for applications.

A flow rule creation/destruction is enqueued on the specified flow queue and offloaded asynchronously to the hardware.



mlx5 PMD updates:

==================

[3] Support the rte_flow changes listed above.

[4] Support sharing HW resources with non-DPDK process using shared ibverbs context and protection domain. Add devargs parameter to map external queues with DPDK indices.



rte_mempool updates:

===================

[5] Optimize the hugepage initialization to speed up the application start-up time.



testpmd updates:

================

[6] Support the changes in rte_flow listed above.




[-- Attachment #2: Type: text/html, Size: 4206 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2021-12-16 16:45 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-16  7:22 NVIDIA roadmap for 22.03 Lior Margalit
2021-12-16 16:45 ` Lior Margalit

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).