| 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
| ||
|
Message-ID: <CAMRc=McT+Q8ZVk9_HTyWd6uS0OoP92E_phwef7CDyDVeNbJCqA@mail.gmail.com>
Date: Fri, 31 Oct 2025 10:00:37 +0100
From: Bartosz Golaszewski <brgl@...ev.pl>
To: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Cc: Linus Walleij <linus.walleij@...aro.org>, Daniel Scally <djrscally@...il.com>,
Heikki Krogerus <heikki.krogerus@...ux.intel.com>,
Sakari Ailus <sakari.ailus@...ux.intel.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>, "Rafael J. Wysocki" <rafael@...nel.org>,
Danilo Krummrich <dakr@...nel.org>, Philipp Zabel <p.zabel@...gutronix.de>,
Krzysztof Kozlowski <krzk@...nel.org>, linux-gpio@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-acpi@...r.kernel.org,
Bartosz Golaszewski <bartosz.golaszewski@...aro.org>
Subject: Re: [PATCH v3 03/10] software node: allow referencing firmware nodes
On Fri, Oct 31, 2025 at 9:24 AM Andy Shevchenko
<andriy.shevchenko@...ux.intel.com> wrote:
>
> On Thu, Oct 30, 2025 at 04:17:48AM -0700, Bartosz Golaszewski wrote:
> > On Thu, 30 Oct 2025 10:41:39 +0100, Andy Shevchenko
> > <andriy.shevchenko@...ux.intel.com> said:
> > > On Wed, Oct 29, 2025 at 01:28:37PM +0100, Bartosz Golaszewski wrote:
> > >>
> > >> At the moment software nodes can only reference other software nodes.
> > >> This is a limitation for devices created, for instance, on the auxiliary
> > >> bus with a dynamic software node attached which cannot reference devices
> > >> the firmware node of which is "real" (as an OF node or otherwise).
> > >>
> > >> Make it possible for a software node to reference all firmware nodes in
> > >> addition to static software nodes. To that end: add a second pointer to
> > >> struct software_node_ref_args of type struct fwnode_handle. The core
> > >> swnode code will first check the swnode pointer and if it's NULL, it
> > >> will assume the fwnode pointer should be set. Rework the helper macros
> > >> and deprecate the existing ones whose names don't indicate the reference
> > >> type.
> > >>
> > >> Software node graphs remain the same, as in: the remote endpoints still
> > >> have to be software nodes.
> > >
> > > ...
> > >
> > >> -#define SOFTWARE_NODE_REFERENCE(_ref_, ...) \
> > >> +#define __SOFTWARE_NODE_REF(_ref, _node, ...) \
> > >> (const struct software_node_ref_args) { \
> > >> - .node = _ref_, \
> > >> + ._node = _ref, \
> > >> .nargs = COUNT_ARGS(__VA_ARGS__), \
> > >> .args = { __VA_ARGS__ }, \
> > >> }
> > >
> > > Okay, looking at this again I think we don't need a new parameter.
> > > We may check the type of _ref_
> > > (actually why are the macro parameters got renamed here and elsewhere?)
> > > and assign the correct one accordingly. I think this is what _Generic()
> > > is good for.
> > >
> >
> > Oh, that's neat, I would love to use _Generic() here but I honest to god have
> > no idea how to make it work. I tried something like:
> >
> > #define __SOFTWARE_NODE_REF(_ref, ...) \
> > _Generic(_ref, \
> > const struct software_node *: \
> > (const struct software_node_ref_args) { \
> > .swnode = _ref, \
> > .nargs = COUNT_ARGS(__VA_ARGS__), \
> > .args = { __VA_ARGS__ }, \
> > }, \
> > struct fwnode_handle *: \
> > (const struct software_node_ref_args) { \
> > .fwnode = _ref, \
> > .nargs = COUNT_ARGS(__VA_ARGS__), \
> > .args = { __VA_ARGS__ }, \
> > } \
> > )
> >
> >
> > But this fails like this:
> >
> > In file included from ./include/linux/acpi.h:16,
> > from drivers/reset/core.c:8:
> > drivers/reset/core.c: In function ‘__reset_add_reset_gpio_device’:
> > drivers/reset/core.c:958:52: error: initialization of ‘const struct
> > software_node *’ from incompatible pointer type ‘struct fwnode_handle
> > *’ [-Wincompatible-pointer-types]
> > 958 | parent->fwnode,
> > | ^~~~~~
> > ./include/linux/property.h:374:35: note: in definition of macro
> > ‘__SOFTWARE_NODE_REF’
> > 374 | .swnode = _ref, \
> >
> > So the right branch is not selected. How exactly would you use it here?
>
> I believe this is an easy task.
>
> But first of all, your series doesn't compile AFAICS:
>
> drivers/reset/core.c:981:6: error: variable 'ret' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized]
> 981 | if (IS_ERR(rgpio_dev->swnode))
> | ^~~~~~~~~~~~~~~~~~~~~~~~~
> drivers/reset/core.c:1001:9: note: uninitialized use occurs here
> 1001 | return ret;
> | ^~~
> drivers/reset/core.c:981:2: note: remove the 'if' if its condition is always false
> 981 | if (IS_ERR(rgpio_dev->swnode))
> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 982 | goto err_put_of_node;
> | ~~~~~~~~~~~~~~~~~~~~
> drivers/reset/core.c:905:13: note: initialize the variable 'ret' to silence this warning
> 905 | int id, ret, lflags;
> | ^
> | = 0
> 1 error generated.
>
You're not wrong but for the record: it builds fine for me with
aarch64-linux-gnu-gcc 14.2 for some reason so I didn't notice it. I'll
fix it.
> So, but to the topic
>
> I have applied this and get the only error as per above
>
> (const struct software_node_ref_args) { \
> - ._node = _ref, \
> + .swnode = _Generic(_ref, const struct software_node *: _ref, default: NULL), \
> + .fwnode = _Generic(_ref, struct fwnode_handle *: _ref, default: NULL), \
>
That works, thanks for the idea.
Bartosz
Powered by blists - more mailing lists