Re: [PATCH v1 1/2] driver core: fw_devlink: Add support for FWNODE_FLAG_BROKEN_PARENT

From: Saravana Kannan <saravanak@google.com>
To: Andrew Lunn <andrew@lunn.ch>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	"Rafael J. Wysocki" <rafael@kernel.org>,
	Linus Walleij <linus.walleij@linaro.org>,
	Vivien Didelot <vivien.didelot@gmail.com>,
	Florian Fainelli <f.fainelli@gmail.com>,
	Vladimir Oltean <olteanv@gmail.com>,
	"David S. Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>, Len Brown <lenb@kernel.org>,
	Alvin Sipraga <ALSI@bang-olufsen.dk>,
	kernel-team@android.com, linux-kernel@vger.kernel.org,
	netdev@vger.kernel.org, linux-acpi@vger.kernel.org
Subject: Re: [PATCH v1 1/2] driver core: fw_devlink: Add support for FWNODE_FLAG_BROKEN_PARENT
Date: Tue, 31 Aug 2021 15:18:35 -0700	[thread overview]
Message-ID: <CAGETcx90dOkw+Yp5ZRNqQq2Ny_ToOKvGJNpvyRohaRQi=SQxhw@mail.gmail.com> (raw)
In-Reply-To: <YS6nxLp5TYCK+mJP@lunn.ch>

On Tue, Aug 31, 2021 at 3:06 PM Andrew Lunn <andrew@lunn.ch> wrote:
>
> On Tue, Aug 31, 2021 at 12:26:30PM -0700, Saravana Kannan wrote:
> > On Tue, Aug 31, 2021 at 6:16 AM Andrew Lunn <andrew@lunn.ch> wrote:
> > >
> > > > > I must admit, my main problem at the moment is -rc1 in two weeks
> > > > > time. It seems like a number of board with Ethernet switches will be
> > > > > broken, that worked before. phy-handle is not limited to switch
> > > > > drivers, it is also used for Ethernet drivers. So it could be, a
> > > > > number of Ethernet drivers are also going to be broken in -rc1?
> > > >
> > > > Again, in those cases, based on your FEC example, fw_devlink=on
> > > > actually improves things.
> > >
> > > Debatable. I did some testing. As expected some boards with Ethernet
> > > switches are now broken.
> >
> > To clarify myself, I'm saying the patch to parse "ethernet" [8] will
> > make things better with fw_devlink=on for FEC. Not sure if you tested
> > that patch or not.
>
> Yes and no. I was tested with the FEC, but because of fw_devlink=on,
> the switches don't probe correctly. So it is not possible to see if it
> helped or not, since its plain broken.
>
> > Not sure what was the tip of the tree with which you bisected.
>
> I manually tested linux-next, v5.14, v5.13 and v5.12 and then
> bisected:
>
> $ git bisect log
> git bisect start
> # good: [9f4ad9e425a1d3b6a34617b8ea226d56a119a717] Linux 5.12
> git bisect good 9f4ad9e425a1d3b6a34617b8ea226d56a119a717
> # bad: [62fb9874f5da54fdb243003b386128037319b219] Linux 5.13
>
> So well away from linux-next which contains the phy-handle parser.
>
> I will try to give the two patches a try today or tomorrow.

If the switches are broken without the phy-handle or ethernet change,
I'm not sure if the "BROKEN_PARENT" patch would help.

>
> > Thanks for testing it out. And worst case, we'll revert phy-handle
> > support.
>
> Which is not enough to fix these Ethernet switches.

Ok, if you can give more specifics on this, I'll look into it. I'd need:
1) The DTS file that you see the issue on.
2) The devices that don't probe when you have the issue (you can find
these in /sys/kernel/debug/devices_deferred)
3) And if possible boot logs with dev_dbg changed to dev_info in
device_link_add() and device_links_check_suppliers()

Thanks,
Saravana