From: Philippe Gerum <rpm@xenomai.org>
To: "François Legal" <devel@thom.fr.eu.org>
Cc: Richard Weinberger <richard.weinberger@gmail.com>,
"Chen, Hongzhan" <hongzhan.chen@intel.com>,
"xenomai@lists.linux.dev" <xenomai@lists.linux.dev>
Subject: Re: Porting RT net driver to dovetail
Date: Mon, 24 Jun 2024 09:46:37 +0200 [thread overview]
Message-ID: <87msnaojrh.fsf@xenomai.org> (raw)
In-Reply-To: <1157-666a8f80-f-41294d00@43403104>
François Legal <devel@thom.fr.eu.org> writes:
> Thanks for the clarification.
>
> I'm not sure I understand the sentence "ATM, we only have a real-time capable raw socket interface".
See https://source.denx.de/Xenomai/xenomai4/libevl/-/blob/master/tidbits/oob-net-icmp.c?ref_type=heads
> In my application, we only use RTNET for level 2 communication. Is this already available.
>
> And so, back to the original question, how do I transition from an RTNET/RTDM NIC to an EVL/??? one ?
>
When the upcoming EVL netstack is fully available (i.e. including the
driver<->evl interface along with the UDP layer), you will add
out-of-band support code to the vanilla driver directly. Typically, a
couple of handlers added to struct net_device_ops, such as:
#ifdef CONFIG_NET_OOB
int (*ndo_enable_oob)(struct net_device *dev);
void (*ndo_disable_oob)(struct net_device *dev);
struct sk_buff * (*ndo_alloc_tx_oob)(struct net_device *ndev);
#endif
The idea is to reuse the vanilla infrastructure as much as possible
instead of forking the driver code. There are a few techniques to have
in mind for doing so while preventing too many merge conflicts whenever
the upstream code changes, the netstack will come with the FEC driver
adapted that way as an illustration.
ETA: end of july.
--
Philippe.
prev parent reply other threads:[~2024-06-24 7:57 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-12 7:07 Porting RT net driver to dovetail François Legal
2024-06-12 8:07 ` Chen, Hongzhan
2024-06-12 12:08 ` François Legal
2024-06-12 12:40 ` Richard Weinberger
2024-06-12 15:01 ` François Legal
2024-06-12 15:09 ` Philippe Gerum
2024-06-13 6:18 ` François Legal
2024-06-24 7:46 ` Philippe Gerum [this message]
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=87msnaojrh.fsf@xenomai.org \
--to=rpm@xenomai.org \
--cc=devel@thom.fr.eu.org \
--cc=hongzhan.chen@intel.com \
--cc=richard.weinberger@gmail.com \
--cc=xenomai@lists.linux.dev \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.