linux-kernel - Re: [PATCH RFT/RFC] mfd: cs42l43: setup true links with software nodes

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAHp75Vf_+WzMn+pmwK4zhLEZtG3Uat4FrdFMCMtPmmX82E3Wzg@mail.gmail.com>
Date: Wed, 19 Nov 2025 16:23:50 +0200
From: Andy Shevchenko <andy.shevchenko@...il.com>
To: Bartosz Golaszewski <brgl@...ev.pl>
Cc: Charles Keepax <ckeepax@...nsource.cirrus.com>, 
	Andy Shevchenko <andriy.shevchenko@...ux.intel.com>, David Rhodes <david.rhodes@...rus.com>, 
	Richard Fitzgerald <rf@...nsource.cirrus.com>, Lee Jones <lee@...nel.org>, 
	Mark Brown <broonie@...nel.org>, Philipp Zabel <p.zabel@...gutronix.de>, 
	Linus Walleij <linus.walleij@...aro.org>, Maciej Strozek <mstrozek@...nsource.cirrus.com>, 
	Andy Shevchenko <andy@...nel.org>, linux-sound@...r.kernel.org, 
	patches@...nsource.cirrus.com, linux-kernel@...r.kernel.org, 
	linux-spi@...r.kernel.org, 
	Bartosz Golaszewski <bartosz.golaszewski@...aro.org>
Subject: Re: [PATCH RFT/RFC] mfd: cs42l43: setup true links with software nodes

On Wed, Nov 19, 2025 at 4:15 PM Bartosz Golaszewski <brgl@...ev.pl> wrote:
> On Wed, Nov 19, 2025 at 3:11 PM Andy Shevchenko
> <andy.shevchenko@...il.com> wrote:
> > On Wed, Nov 19, 2025 at 3:35 PM Bartosz Golaszewski <brgl@...ev.pl> wrote:
> > > On Wed, Nov 19, 2025 at 2:27 PM Charles Keepax
> > > <ckeepax@...nsource.cirrus.com> wrote:
> > > > On Wed, Nov 19, 2025 at 02:07:55PM +0100, Bartosz Golaszewski wrote:
> > > > > On Wed, Nov 19, 2025 at 1:53 PM Charles Keepax
> > > > > <ckeepax@...nsource.cirrus.com> wrote:

...

> > > > > I still don't understand what the software node that's already
> > > > > assigned to the SPI device is though? device_add_software_node()
> > > > > should work just fine if the only other firmware node the device has
> > > > > is the ACPI device node.
> > > >
> > > > Its the software node we assigned to the first MFD cell, that one
> > > > succeeds but attaches itself to the ACPI node as a secondary.
> > > > When we get to the second cell we try to attach a new node but we
> > > > get the one from the first cell since they share an ACPI node.
> > >
> > > Ah, I see now. That's indeed a fundamental problem that can't be
> > > easily solved. Andy is right about the needed change but phew that'll
> > > be a big one...
> > >
> > > > I think as Andy pointed out though the first 4 patches in your
> > > > chain do loosely want we want. Previously, we used the name to
> > > > point to the actual pinctrl driver, your patches should let us
> > > > do that properly through the fwnode. So we can drop the pinctrl
> > > > swnode and just have the cs-gpios bit point at the actual fwnode
> > > > instead. I am trying to hack together a strawman but its failing
> > > > in a lightly odd way. Hopefully I can get that sorted fairly
> > > > soon and post, or I guess I could post a version earlier if you
> > > > wanted a look in the knowledge it still doesn't work?
> > >
> > > Yeah, I'd like that. I want to get it upstream and have interest in
> > > getting it fixed ASAP.
> >
> > I also have a side interest that the reset-gpio becomes an agnostic driver.
> >
> > > For the record: I believe the logic behind this patch is the correct
> > > approach. It uses the existing MFD infrastructure that's there for
> > > exactly this reason. However without being able to have an arbitrary
> > > number of firmware nodes attached to a device, that'll be impossible.
> > >
> > > Just an idea: we could try to do the conversion gradually - by first
> > > adding that list of fwnodes to struct device in parallel to the
> > > current approach of having the secondary pointer and then go from
> > > there step by step.
> >
> > My idea was to mark the fwnode with __private and fix the (ab)users,
> > should not be so many. Can somebody mock up a coccinelle script to
> > find all dereferences of fwnode from struct device?
>
> I think you're underestimating the level of complexity. What about the
> concept of dev_fwnode()? It literally makes no sense if we switch to a
> list of fwnodes.

Why not? It will return the pointer to the primary node. You can look
for example how the list of the DMA descriptors is done in
drivers/dma/dw/core.c. Not the best solution, but gives you an idea of
how it may look.

> For it to make sense we'd have to have a kind of "dynamic" firmware
> node attached to a device which we'd fill with an aggregation of all
> properties from firmware nodes in the list.

"Dynamic" is just a node in the list. The only potential problem here
is prioritisation. Should we add to the head, tail or insert? But
converting current approach will be straightforward.

-- 
With Best Regards,
Andy Shevchenko