Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:2 as permitted sender) client-ip=2620:137:e000::3:2;
Date:   Wed, 6 Dec 2023 18:43:29 +0200
From:   Vladimir Oltean <olteanv@gmail.com>
To:     Oleksij Rempel <o.rempel@pengutronix.de>
Cc:     Paolo Abeni <pabeni@redhat.com>,
        "David S. Miller" <davem@davemloft.net>,
        Andrew Lunn <andrew@lunn.ch>,
        Eric Dumazet <edumazet@google.com>,
        Florian Fainelli <f.fainelli@gmail.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Woojung Huh <woojung.huh@microchip.com>,
        Arun Ramadoss <arun.ramadoss@microchip.com>,
        kernel@pengutronix.de, linux-kernel@vger.kernel.org,
        netdev@vger.kernel.org, UNGLinuxDriver@microchip.com
Subject: Re: [PATCH net-next v1 3/3] net: dsa: microchip: Fix PHY loopback
 configuration for KSZ8794 and KSZ8873
Message-ID: <20231206164329.fszkmpilktyq6r6v@skbuf>
References: <20231121152426.4188456-1-o.rempel@pengutronix.de>
 <20231121152426.4188456-3-o.rempel@pengutronix.de>
 <35045f6ef6a5b274063186c065a8215088b94cd5.camel@redhat.com>
 <20231206085520.GA1293736@pengutronix.de>
 <20231206151406.75eglqtsrrb4vegf@skbuf>
 <20231206155440.GA1324895@pengutronix.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20231206155440.GA1324895@pengutronix.de>
Precedence: bulk

On Wed, Dec 06, 2023 at 04:54:40PM +0100, Oleksij Rempel wrote:
> > I don't see DSA implementing ndo_set_features(), nor offering NETIF_F_LOOPBACK.
> > The PHY is integrated, so DSA is the only relevant netdev driver. Is
> > there any other way to test this functionality?
> 
> yes - net_selftest()

Ok, I didn't notice net_test_phy_loopback_enable(). So it can be
triggered after all, it seems.

But I mean, if it's exclusively a selftest that fails, and has always
failed since its introduction, I think it can be considered new
development work when it stops failing? I don't believe that the impact
of the bug is relevant for users. It's not a production functionality.
Documentation/process/stable-kernel-rules.rst doesn't specifically say
this, but it does imply that we should triage the "real bugs that bother
people" as much as possible.

> > If not, I think it's a case of "tree falling in the woods and nobody
> > hearing it". Not "stable" material. But it definitely has nothing to do
> > with not caring about the switch variant.
> 
> Sorry, my intention is not to criticize anyone. I am not getting
> feedbacks or bug reports for ksz88xx variants, so it seems like not many
> people use it in upstream.
> 
> When I have time slots to work on this driver, I try to use them to do
> fixes and also clean up the code. Since there is some sort of fog of
> uncertainty about when I get the next time slot, or even if I get it at
> all, I am trying to push both fixes and cleanups together.
> 
> But, you are right, it is not a good reason for not caring about stable :)
> 
> What is the decision about this patch set?

I wouldn't bend over backwards for this, and reorder the patches.
I would spend my time doing more meaningful things.