linux-kernel - [PATCH 0/3] gpio: switch to device_for_each_chilld_node

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-next>] [day] [month] [year] [list]

Message-Id: <20240928-gpio_device_for_each_child_node_scoped-v1-0-c20eff315f4f@gmail.com>
Date: Sat, 28 Sep 2024 21:47:34 +0200
From: Javier Carrasco <javier.carrasco.cruz@...il.com>
To: Hoan Tran <hoan@...amperecomputing.com>, 
 Serge Semin <fancer.lancer@...il.com>, 
 Linus Walleij <linus.walleij@...aro.org>, 
 Bartosz Golaszewski <brgl@...ev.pl>, 
 Mika Westerberg <mika.westerberg@...ux.intel.com>, 
 Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Cc: linux-gpio@...r.kernel.org, linux-kernel@...r.kernel.org, 
 linux-acpi@...r.kernel.org, 
 Javier Carrasco <javier.carrasco.cruz@...il.com>
Subject: [PATCH 0/3] gpio: switch to device_for_each_chilld_node_scoped()

This series switches from the device_for_each_child_node() macro to its
scoped variant. This makes the code more robust if new early exits are
added to the loops, because there is no need for explicit calls to
fwnode_handle_put(), which also simplifies existing code.

The non-scoped macros to walk over nodes turn error-prone as soon as
the loop contains early exits (break, goto, return), and patches to
fix them show up regularly, sometimes due to new error paths in an
existing loop [1].

The uses of device_for_each_child_node() with no early exits have been
left untouched because their simpilicty justifies the non-scoped
variant. In particular for gpio, there is a single case in gpio-hisi.c
where the loop does not have early exists. But if desired, it can be
easily converted as well to enforce the usage of the scoped variant when
"borrowing" existing code.

Note that the child node is now declared in the macro, and therefore the
explicit declaration is no longer required.

The general functionality should not be affected by this modification.
Apart from automatically decrementing the node's refcount when it goes
out of scope, the scoped variant works exactly the same as the
non-scoped.

If functional changes are found, please report them back as errors.

Just in case, and even though the scoped macro already has multiple
users, I carried out a quick test with gpio-sim and a few nodes defined
in a devicetree, and everything works as expected.

Link:
https://lore.kernel.org/all/20240901160829.709296395@linuxfoundation.org/
[1]

Signed-off-by: Javier Carrasco <javier.carrasco.cruz@...il.com>
---
Javier Carrasco (3):
      gpio: dwapb: switch to device_for_each_child_node_scoped()
      gpio: sim: switch to device_for_each_child_node_scoped()
      gpio: acpi: switch to device_for_each_child_node_scoped()

 drivers/gpio/gpio-dwapb.c   | 4 +---
 drivers/gpio/gpio-sim.c     | 7 ++-----
 drivers/gpio/gpiolib-acpi.c | 4 +---
 3 files changed, 4 insertions(+), 11 deletions(-)
---
base-commit: 40e0c9d414f57d450e3ad03c12765e797fc3fede
change-id: 20240927-gpio_device_for_each_child_node_scoped-a2071fe5c8c6

Best regards,
-- 
Javier Carrasco <javier.carrasco.cruz@...il.com>