From: Claude Code Review Bot <claude-review@example.com>
To: dri-devel-reviews@example.com
Subject: Claude review: platform/surface: gpe: use platform_device_register_full()
Date: Sat, 16 May 2026 11:52:19 +1000 [thread overview]
Message-ID: <review-patch2-20260513-swnode-remove-on-dev-unreg-v6-2-f9c58939df27@oss.qualcomm.com> (raw)
In-Reply-To: <20260513-swnode-remove-on-dev-unreg-v6-2-f9c58939df27@oss.qualcomm.com>
Patch Review
**Bug: `surface_gpe_exit()` is not updated and will use-after-free.**
The init conversion is clean -- replacing manual `fwnode_create_software_node()` + `platform_device_alloc()` + `platform_device_add()` with a single `platform_device_register_full()` using `.properties`:
```c
pdevinfo = (struct platform_device_info){
.name = "surface_gpe",
.id = PLATFORM_DEVID_NONE,
.properties = match->driver_data,
};
pdev = platform_device_register_full(&pdevinfo);
```
However, the existing `surface_gpe_exit()` (unchanged by this patch) still does:
```c
static void __exit surface_gpe_exit(void)
{
struct fwnode_handle *fwnode = surface_gpe_device->dev.fwnode;
platform_device_unregister(surface_gpe_device);
platform_driver_unregister(&surface_gpe_driver);
fwnode_remove_software_node(fwnode);
}
```
With the old code, the fwnode was a standalone (non-managed) software node that persisted after `platform_device_unregister()`. With the new code, `platform_device_register_full()` with `.properties` calls `device_create_managed_software_node()`, which creates a **managed** node. During `platform_device_unregister()` -> `device_del()` -> `software_node_notify_remove()`, the managed node is cleaned up and its `kobject` refcount can reach zero, freeing the `swnode`. The subsequent `fwnode_remove_software_node(fwnode)` then dereferences freed memory.
**Fix needed**: The exit function should be updated to remove the `fwnode_remove_software_node()` call and the saved `fwnode` pointer:
```c
static void __exit surface_gpe_exit(void)
{
platform_device_unregister(surface_gpe_device);
platform_driver_unregister(&surface_gpe_driver);
}
```
---
---
Generated by Claude Code Patch Reviewer
next prev parent reply other threads:[~2026-05-16 1:52 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-13 15:04 [PATCH v6 0/6] driver core: remove software node from platform devices on device release Bartosz Golaszewski
2026-05-13 15:04 ` [PATCH v6 1/6] driver core: platform: remove software node on release() Bartosz Golaszewski
2026-05-16 1:52 ` Claude review: " Claude Code Review Bot
2026-05-13 15:04 ` [PATCH v6 2/6] platform/surface: gpe: use platform_device_register_full() Bartosz Golaszewski
2026-05-16 1:52 ` Claude Code Review Bot [this message]
2026-05-13 15:04 ` [PATCH v6 3/6] drm/xe/i2c: use device_create_managed_software_node() Bartosz Golaszewski
2026-05-16 1:52 ` Claude review: " Claude Code Review Bot
2026-05-13 15:04 ` [PATCH v6 4/6] driver core: platform: unify release path Bartosz Golaszewski
2026-05-16 1:52 ` Claude review: " Claude Code Review Bot
2026-05-13 15:04 ` [PATCH v6 5/6] kunit: provide kunit_software_node_register() Bartosz Golaszewski
2026-05-16 1:52 ` Claude review: " Claude Code Review Bot
2026-05-13 15:04 ` [PATCH v6 6/6] driver core: platform: tests: add test cases for correct swnode removal Bartosz Golaszewski
2026-05-16 1:52 ` Claude review: " Claude Code Review Bot
2026-05-14 0:24 ` (subset) [PATCH v6 0/6] driver core: remove software node from platform devices on device release Danilo Krummrich
2026-05-16 1:52 ` Claude review: " Claude Code Review Bot
-- strict thread matches above, loose matches on Subject: below --
2026-05-12 11:59 [PATCH v5 0/7] " Bartosz Golaszewski
2026-05-12 11:59 ` [PATCH v5 1/7] platform/surface: gpe: use platform_device_register_full() Bartosz Golaszewski
2026-05-16 3:28 ` Claude review: " Claude Code Review Bot
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=review-patch2-20260513-swnode-remove-on-dev-unreg-v6-2-f9c58939df27@oss.qualcomm.com \
--to=claude-review@example.com \
--cc=dri-devel-reviews@example.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox