From: Niklas Schnelle <schnelle@linux.ibm.com>
To: Parav Pandit <parav@nvidia.com>, "David S.Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org, Stefan Raspl <raspl@linux.ibm.com>,
Viktor Mihajlovski <mihajlov@linux.ibm.com>
Subject: Re: Interface naming on v5.14-rc1 regressed, possible unintended revert of b28d8f0c25a9?
Date: Mon, 19 Jul 2021 12:49:38 +0200 [thread overview]
Message-ID: <3d6005401d49c2a1f3303cf9aa3d672c0cb5ce74.camel@linux.ibm.com> (raw)
In-Reply-To: <PH0PR12MB54813D21F0F3338CD7764DE3DC149@PH0PR12MB5481.namprd12.prod.outlook.com>
On Tue, 2021-07-13 at 16:13 +0000, Parav Pandit wrote:
> Hi Niklas,
>
> > From: Niklas Schnelle <schnelle@linux.ibm.com>
> > Sent: Tuesday, July 13, 2021 9:18 PM
> >
> > Hi Parav, Hi Dave,
> >
> > With v5.13 you added commit b28d8f0c25a9 ("devlink: Correct VIRTUAL port
> > to not have phys_port attributes") which changed the interface names of
> > VFs from e.g. ens2f0np0v0 to ens2f0v0. I did see that change for the v5.13-rc
> > cycle.
> >
> > Now I just noticed that with v5.14-rc1 the old name is back. Looking at the
> > above commit and the code in v5.14-rc1 it looks to me like that commit was
> > partially reverted. In particular, the return of -EOPNOTSUPP from
> > __devlink_port_phys_port_name_get() in case of port flavour "virtual" is
> > gone while the part touching devlink_nl_port_attrs_put() seems to be there.
> >
> > Now for some reason looking at git blame I still see the original commit
> > acf1ee44ca5da ("devlink: Introduce devlink port flavour
> > virtual") for the non removed lines. Could it be that somehow this got lost
> > while resolving a merge conflict?
> >
>
> Thanks for the bisecting and reporting.
> I searched further.
> I found that, it got introduced by below merge commit.
> but I haven’t yet found the patch that removed the return of -EOPNOTSUPP.
>
> commit 126285651b7f95282a0afe3a1b0221419b31d989
> Merge: 9977d6f56bac 3822d0670c9d
> Author: David S. Miller <davem@davemloft.net>
> Date: Mon Jun 7 13:01:52 2021 -0700
>
> Merge ra.kernel.org:/pub/scm/linux/kernel/git/netdev/net
>
> Bug fixes overlapping feature additions and refactoring, mostly.
>
> Signed-off-by: David S. Miller <davem@davemloft.net>
Just a FYI, I still see interface names with '..np0' suffix on v5.14-
rc2. Any progress on tracking down what happened? Also I think I
accidentially used the wrong mail address for the netdev list before,
sorry about that.
parent reply other threads:[~2021-07-19 10:49 UTC|newest]
Thread overview: expand[flat|nested] mbox.gz Atom feed
[parent not found: <PH0PR12MB54813D21F0F3338CD7764DE3DC149@PH0PR12MB5481.namprd12.prod.outlook.com>]
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=3d6005401d49c2a1f3303cf9aa3d672c0cb5ce74.camel@linux.ibm.com \
--to=schnelle@linux.ibm.com \
--cc=davem@davemloft.net \
--cc=mihajlov@linux.ibm.com \
--cc=netdev@vger.kernel.org \
--cc=parav@nvidia.com \
--cc=raspl@linux.ibm.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.