Re: Dpdk packet reception

[Rajasekhar Pulluru <pullururajasekhar@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1637 bytes --]

Hi Alexander,

Glad to see the response from the community on this thread.

Thanks for the information, I will go through the code that you pointed
out. Does Intel(R) Atom(TM) CPU C37xx support the optimized power state
mode? Is there a chance of losing packets(rx-miss in NIC) in the low power
mode?

Thanks & Regards,
Rajasekhar
On Mon, Mar 13, 2023 at 9:44 PM Alexander Kozyrev <akozyrev@nvidia.com>
wrote:

> Hi Rajasekhar, Power Management API can “put the CPU into optimized power
> state and monitor the Ethernet PMD RX descriptor address, waking the CPU up
> whenever there’s new traffic”.
>
> Please take a look at the rte_power_monitor() and dpdk-l3fwd-power
> application example to see if it works for you. What platform do you use?
> It is only supported on recent Intel CPUs and ARMs.
>
>
>
> Regards,
>
> Alex
>
>
>
> *From:* Rajasekhar Pulluru <pullururajasekhar@gmail.com>
> *Sent:* March 10, 2023 15:57
> *To:* dev@dpdk.org
> *Subject:* Dpdk packet reception
>
>
>
> Hi Team,
>
>
>
> Would like to know if dpdk has a select(fd)/epoll(fd) like mechanism to
> wake-up the application (say rx-pthread) asynchronously upon packet
> reception without enabling rxq interrupts?
>
>
>
> I can't afford to lose a cpu-core polling continuously to check for
> packets using rte_eth_rx_burst.And I only have 1 core to receive packets
> from multiple ports.I don't want to use usleep of fixed timeout between
> rte_eth_rx_burst api invocation, as it potentially could lead to packets
> being rx-missed in 1 or more ports.
>
>
>
> Thanks & Regards,
>
> Rajasekhar
>

[-- Attachment #2: Type: text/html, Size: 3344 bytes --]