All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ido Schimmel <idosch@nvidia.com>
To: Jakub Kicinski <kuba@kernel.org>
Cc: netdev@vger.kernel.org, davem@davemloft.net, pabeni@redhat.com,
	edumazet@google.com, andrew+netdev@lunn.ch, horms@kernel.org,
	petrm@nvidia.com
Subject: Re: [PATCH net-next 1/3] net: Do not unconditionally turn on carrier when clearing protodown
Date: Sun, 3 May 2026 21:08:03 +0300	[thread overview]
Message-ID: <20260503180803.GA269092@shredder> (raw)
In-Reply-To: <20260501180843.069635aa@kernel.org>

On Fri, May 01, 2026 at 06:08:43PM -0700, Jakub Kicinski wrote:
> On Wed, 29 Apr 2026 15:46:22 +0300 Ido Schimmel wrote:
> > Solve this by adding a new NDO that allows different drivers to react
> > differently to protodown being cleared. Change the core to invoke the
> > NDO and implement it for vxlan and macvlan. Maintain the current
> > behavior for the former, but for macvlan implement the NDO by
> > transferring the operational state from its lower device.
> 
> Maybe just me but an NDO for this seems like an overkill.
> Isn't our range of options for the callback either follow
> lower or carrier on? Do you expect more complex logic?

Thanks for reviewing. I don't expect more complex logic, but resolving
the linked device via the driver is more straightforward (unless I'm
missing something).

There are drivers that only implement ndo_get_iflink(), but don't
implement get_link_net(), so it's unclear in which netns we need to look
up the device using the ifindex we got from ndo_get_iflink(). Netdevsim
is one example. There are also some drivers that only implement
get_link_net() such as vxlan.

The popular drivers such as vlan, macvlan and veth obviously implement
both.

I added a helper [1] that resolves the linked device if these operations
are present and otherwise returns the device itself. Similar to
dev_get_iflink() which only gives us the ifindex.

Another option is to add a new NDO that gives us a pointer to the linked
device.

[1] https://github.com/idosch/linux/commit/34c57bceb1a7e52da0f973914f8a4dd08b7f9072.patch

  reply	other threads:[~2026-05-03 18:08 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-04-29 12:46 [PATCH net-next 0/3] net: Fix protodown with macvlan Ido Schimmel
2026-04-29 12:46 ` [PATCH net-next 1/3] net: Do not unconditionally turn on carrier when clearing protodown Ido Schimmel
2026-05-02  1:08   ` Jakub Kicinski
2026-05-03 18:08     ` Ido Schimmel [this message]
2026-05-05  0:55       ` Jakub Kicinski
2026-04-29 12:46 ` [PATCH net-next 2/3] macvlan: Do not transfer operational state when protodown is enabled Ido Schimmel
2026-05-02  1:09   ` Jakub Kicinski
2026-05-03 18:19     ` Ido Schimmel
2026-05-05  0:57       ` Jakub Kicinski
2026-04-29 12:46 ` [PATCH net-next 3/3] selftests: net: Add protodown tests Ido Schimmel

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=20260503180803.GA269092@shredder \
    --to=idosch@nvidia.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=horms@kernel.org \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=petrm@nvidia.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 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.