Re: [PATCH v3 net-next] bonding: pair enable_port with slave_arr_updates

[Jay Vosburgh <jay.vosburgh@canonical.com>]

Jakub Kicinski <kuba@kernel.org> wrote:
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

>On Sun, 6 Feb 2022 21:52:11 -0800 Mahesh Bandewar (=E0=A4=AE=E0=A4=B9=E0=
=A5=87=E0=A4=B6 =E0=A4=AC=E0=A4=82=E0=A4=A1=E0=A5=87=E0=A4=B5=E0=A4=BE=E0=
=A4=B0) wrote:
>> On Fri, Feb 4, 2022 at 7:59 PM Jakub Kicinski <kuba@kernel.org> wrote:
>> > Quacks like a fix, no? It's tagged for net-next and no fixes tag,
>> > is there a reason why?=20=20
>>=20
>> Though this fixes some corner cases, I couldn't find anything obvious
>> that I can report as "fixes" hence decided otherwise. Does that make
>> sense?
>
>So it's was not introduced in the refactorings which added
>update_slave_arr? If the problem existed forever we can put:
>
>Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
>
>it's just an indication how far back the backporting should go.
>For anything older than oldest LTS (4.9) the exact tag probably
>doesn't matter all that much.

	I think the correct Fixes line would be the commit that
introduces the array logic in the first place, which I believe is:

Fixes: ee6377147409 ("bonding: Simplify the xmit function for modes that us=
e xmit_hash")

	This dates to 3.18.

	-J

---
	-Jay Vosburgh, jay.vosburgh@canonical.com