Go to file
Elena Agostini c7ebd65c13 gpudev: add communication list
In heterogeneous computing system, processing is not only in the CPU.
Some tasks can be delegated to devices working in parallel.
When mixing network activity with task processing there may be the need
to put in communication the CPU with the device in order to synchronize
operations.

An example could be a receive-and-process application
where CPU is responsible for receiving packets in multiple mbufs
and the GPU is responsible for processing the content of those packets.

The purpose of this list is to provide a buffer in CPU memory visible
from the GPU that can be treated as a circular buffer
to let the CPU provide fondamental info of received packets to the GPU.

A possible use-case is described below.

CPU:
- Trigger some task on the GPU
- in a loop:
    - receive a number of packets
    - provide packets info to the GPU

GPU:
- Do some pre-processing
- Wait to receive a new set of packet to be processed

Layout of a communication list would be:

     -------
    |   0    | => pkt_list
    | status |
    | #pkts  |
     -------
    |   1    | => pkt_list
    | status |
    | #pkts  |
     -------
    |   2    | => pkt_list
    | status |
    | #pkts  |
     -------
    |  ....  | => pkt_list
     -------

Signed-off-by: Elena Agostini <eagostini@nvidia.com>
2021-11-08 17:20:53 +01:00
.ci ci: fix aarch64 cross compilation in GHA 2021-11-01 18:15:24 +01:00
.github/workflows ci: fix aarch64 cross compilation in GHA 2021-11-01 18:15:24 +01:00
app gpudev: add communication list 2021-11-08 17:20:53 +01:00
buildtools buildtools: fix build with Meson 0.60 2021-11-03 14:16:29 +01:00
config config/arm: split aarch32 options 2021-11-08 09:17:06 +01:00
devtools devtools: forbid indent with tabs in Meson 2021-11-02 19:25:30 +01:00
doc gpudev: add communication list 2021-11-08 17:20:53 +01:00
drivers gpudev: introduce GPU device class library 2021-11-08 17:20:52 +01:00
examples crypto/mlx5: support 1MB data-unit 2021-11-04 19:46:27 +01:00
kernel kni: update kernel API to set random MAC address 2021-11-08 11:51:39 +01:00
lib gpudev: add communication list 2021-11-08 17:20:53 +01:00
license lib: remove librte_ prefix from directory names 2021-04-21 14:04:09 +02:00
usertools usertools/devbind: add Kunpeng DMA 2021-11-07 20:04:28 +01:00
.editorconfig doc: add Meson coding style to contributors guide 2021-04-21 14:04:09 +02:00
.gitattributes improve git diff 2016-11-13 15:25:12 +01:00
.gitignore gpudev: introduce GPU device class library 2021-11-08 17:20:52 +01:00
.travis.yml version: 21.11-rc0 2021-08-17 08:37:52 +02:00
ABI_VERSION version: 21.11-rc0 2021-08-17 08:37:52 +02:00
MAINTAINERS gpudev: introduce GPU device class library 2021-11-08 17:20:52 +01:00
Makefile build: create dummy Makefile 2020-09-07 23:51:57 +02:00
meson_options.txt config: add option for atomic mbuf reference counting 2021-10-25 17:53:30 +02:00
meson.build lib: remove librte_ prefix from directory names 2021-04-21 14:04:09 +02:00
README license: introduce SPDX identifiers 2018-01-04 22:41:38 +01:00
VERSION version: 21.11-rc1 2021-10-25 22:42:47 +02:00

DPDK is a set of libraries and drivers for fast packet processing.
It supports many processor architectures and both FreeBSD and Linux.

The DPDK uses the Open Source BSD-3-Clause license for the core libraries
and drivers. The kernel components are GPL-2.0 licensed.

Please check the doc directory for release notes,
API documentation, and sample application information.

For questions and usage discussions, subscribe to: users@dpdk.org
Report bugs and issues to the development mailing list: dev@dpdk.org