From: Chaoyong He <chaoyong.he@corigine.com>
To: Thomas Monjalon <thomas@monjalon.net>,
Ferruh Yigit <ferruh.yigit@xilinx.com>,
"dev@dpdk.org" <dev@dpdk.org>
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 07:42:07 +0000 [thread overview]
Message-ID: <PH0PR13MB556870BFE1DDC936068F68119E4F9@PH0PR13MB5568.namprd13.prod.outlook.com> (raw)
In-Reply-To: <2126161.Icojqenx9y@thomas>
> 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.
>
Actually, in the v1--v5 of this patch series, we did create a control port which is not
in the rte_eth_devices[] array, so it won't exposed to the application.
>
> > > 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 7:42 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
2022-09-21 7:42 ` Chaoyong He [this message]
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=PH0PR13MB556870BFE1DDC936068F68119E4F9@PH0PR13MB5568.namprd13.prod.outlook.com \
--to=chaoyong.he@corigine.com \
--cc=andrew.rybchenko@oktetlabs.ru \
--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 \
--cc=thomas@monjalon.net \
/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).