From: Thomas Monjalon <thomas@monjalon.net>
To: Ferruh Yigit <ferruh.yigit@xilinx.com>,
"dev@dpdk.org" <dev@dpdk.org>,
Chaoyong He <chaoyong.he@corigine.com>
Cc: oss-drivers <oss-drivers@corigine.com>,
Niklas Soderlund <niklas.soderlund@corigine.com>,
Jerin Jacob Kollanukkaran <jerinj@marvell.com>,
Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>,
David Marchand <david.marchand@redhat.com>
Subject: Re: [PATCH v9 07/12] net/nfp: add flower ctrl VNIC related logics
Date: Wed, 21 Sep 2022 09:29:43 +0200 [thread overview]
Message-ID: <2126161.Icojqenx9y@thomas> (raw)
In-Reply-To: <SJ0PR13MB5545CC1493FDE4CB4F8EC00C9E4F9@SJ0PR13MB5545.namprd13.prod.outlook.com>
21/09/2022 04:02, Chaoyong He:
> > On 9/15/2022 11:44 AM, Chaoyong He wrote:
> > > Adds the setup/start logic for the ctrl vNIC. This vNIC is used by the
> > > PMD and flower firmware application as a communication channel
> > between
> > > driver and firmware. In the case of OVS it is also used to communicate
> > > flow statistics from hardware to the driver.
> > >
> > > A rte_eth device is not exposed to DPDK for this vNIC as it is
> > > strictly used internally by flower logic.
> > >
> >
> > Hi Chaoyong,
> >
> > Similar comment with previous versions, interface is created using regular
> > 'rte_eth_dev_allocate()' API, I think interface will be visible to application, I
> > can't understand the need of creating an interface for control.
You didn't reply to this.
Why the control port should be exposed to the application?
We recommend not using ethdev for this.
> > What is the communication method between driver and FW?
> > Since one of the following patches (09/12) introduces Rx/Tx for ctrl interface,
> > is device interface is control packets (similar to network data packets)?
> >
>
> Basically, the 'control message' is exist in the form of normal data packets.
>
> When we use the flower firmware application, there exist two types of packets for now,
> and they are identified only from the prepend meta-data.
>
> Bit 3 3 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0
> -----\ 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0
> Word +---------------+---------------+---------------+---------------+
> 0 | type | type | type | type |
> +---------------+---------------+---------------+---------------+
> The 'control message' packets are processed by the ctrl vNIC.
> The 'normal' packets are processed by the pf vNIC.
>
> The communication method between driver and firmware is decided by the
> designment of hardware and firmware.
>
> The kernel driver also has the same ctrl vNIC and pf vNIC ethdev and the usage is same.
>
> > > Because of the add of ctrl vNIC, a new PCItoCPPBar is needed. Modify
> > > the related logics.
> > >
> > > Signed-off-by: Chaoyong He <chaoyong.he@corigine.com>
> > > Reviewed-by: Niklas Söderlund <niklas.soderlund@corigine.com>
next prev parent reply other threads:[~2022-09-21 7:29 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-15 10:44 [PATCH v9 00/12] preparation for the rte_flow offload of nfp PMD Chaoyong He
2022-09-15 10:44 ` [PATCH v9 01/12] net/nfp: move app specific attributes to own struct Chaoyong He
2022-09-15 10:44 ` [PATCH v9 02/12] net/nfp: simplify initialization and remove dead code Chaoyong He
2022-09-15 10:44 ` [PATCH v9 03/12] net/nfp: move app specific init logic to own function Chaoyong He
2022-09-15 10:44 ` [PATCH v9 04/12] net/nfp: add initial flower firmware support Chaoyong He
2022-09-15 10:44 ` [PATCH v9 05/12] net/nfp: add flower PF setup logic Chaoyong He
2022-09-20 14:57 ` Ferruh Yigit
2022-09-21 2:50 ` Chaoyong He
2022-09-21 7:35 ` Thomas Monjalon
2022-09-21 7:47 ` Chaoyong He
2022-09-15 10:44 ` [PATCH v9 06/12] net/nfp: add flower PF related routines Chaoyong He
2022-09-15 10:44 ` [PATCH v9 07/12] net/nfp: add flower ctrl VNIC related logics Chaoyong He
2022-09-20 14:56 ` Ferruh Yigit
2022-09-21 2:02 ` Chaoyong He
2022-09-21 7:29 ` Thomas Monjalon [this message]
2022-09-21 7:42 ` Chaoyong He
2022-09-15 10:44 ` [PATCH v9 08/12] net/nfp: move common rxtx function for flower use Chaoyong He
2022-09-15 10:44 ` [PATCH v9 09/12] net/nfp: add flower ctrl VNIC rxtx logic Chaoyong He
2022-09-15 10:44 ` [PATCH v9 10/12] net/nfp: add flower representor framework Chaoyong He
2022-09-15 10:44 ` [PATCH v9 11/12] net/nfp: move rxtx function to header file Chaoyong He
2022-09-15 10:44 ` [PATCH v9 12/12] net/nfp: add flower PF rxtx logic Chaoyong He
2022-09-20 14:56 ` [PATCH v9 00/12] preparation for the rte_flow offload of nfp PMD Ferruh Yigit
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=2126161.Icojqenx9y@thomas \
--to=thomas@monjalon.net \
--cc=andrew.rybchenko@oktetlabs.ru \
--cc=chaoyong.he@corigine.com \
--cc=david.marchand@redhat.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@xilinx.com \
--cc=jerinj@marvell.com \
--cc=niklas.soderlund@corigine.com \
--cc=oss-drivers@corigine.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).