Devicetree
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Shivendra Pratap" <shivendra.pratap@oss.qualcomm.com>
Cc: robh@kernel.org, conor+dt@kernel.org, devicetree@vger.kernel.org,
	krzk+dt@kernel.org
Subject: Re: [PATCH v22 08/13] mfd: core: Add firmware-node support to MFD cells
Date: Thu, 14 May 2026 23:19:42 +0000	[thread overview]
Message-ID: <20260514231943.93B30C2BCB3@smtp.kernel.org> (raw)
In-Reply-To: <20260514-arm-psci-system_reset2-vendor-reboots-v22-8-28a5bde07483@oss.qualcomm.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Use-After-Free due to incorrect lifecycle management of non-OF fwnodes using devm_add_action.
--

commit 346fbcb3dbf6a1ca28eec9135059c21e18423182
Author: Shivendra Pratap <shivendra.pratap@oss.qualcomm.com>

mfd: core: Add firmware-node support to MFD cells

This commit extends struct mfd_cell with a callback to allow drivers to
provide an explicit firmware node. The node is assigned to the MFD child
device during registration if one is not assigned by other matching
mechanisms.

> diff --git a/drivers/mfd/mfd-core.c b/drivers/mfd/mfd-core.c
> --- a/drivers/mfd/mfd-core.c
> +++ b/drivers/mfd/mfd-core.c

[ ... ]

> @@ -224,6 +231,29 @@ static int mfd_add_device(struct device *parent, int id,

[ ... ]

> +			/*
> +			 * platform_device_release() drops only of_node refs.
> +			 * Track non-OF fwnodes explicitly so they are put on
> +			 * all teardown paths.
> +			 */
> +			if (!to_of_node(fwnode)) {
> +				ret = devm_add_action(&pdev->dev,
> +						      mfd_child_fwnode_put,
> +						      fwnode);

Is it safe to tie the fwnode lifetime to the driver bind and unbind
lifecycle using devm_add_action() here?

Since devres managed resources are released when a driver unbinds, if a
driver binds to this MFD child device and is later unbound via sysfs,
device_unbind_cleanup() will call devres_release_all(). This triggers
mfd_child_fwnode_put() and drops the reference, freeing the underlying memory
while the device still has its fwnode pointer populated.

Could a subsequent driver rebind trigger a use-after-free when accessing
the dropped fwnode?

Also, during device removal, device_del() calls devres_release_all() before
emitting the KOBJ_REMOVE uevent. The platform bus uevent handler then accesses
the fwnode:

device_del()
   devres_release_all(dev)
   kobject_uevent(&dev->kobj, KOBJ_REMOVE)
      platform_uevent()
         acpi_device_uevent_modalias()
            ACPI_COMPANION(dev)

Will evaluating ACPI_COMPANION() access dev->fwnode after it has already
been freed by devres_release_all(), causing a use-after-free during
standard device teardown?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260514-arm-psci-system_reset2-vendor-reboots-v22-0-28a5bde07483@oss.qualcomm.com?part=8

  reply	other threads:[~2026-05-14 23:19 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-14 14:25 [PATCH v22 00/13] Implement PSCI reboot mode driver for PSCI resets Shivendra Pratap
2026-05-14 14:25 ` [PATCH v22 01/13] power: reset: reboot-mode: Remove devres based allocations Shivendra Pratap
2026-05-14 14:25 ` [PATCH v22 02/13] power: reset: reboot-mode: Add support for 64 bit magic Shivendra Pratap
2026-05-14 20:15   ` sashiko-bot
2026-05-14 14:25 ` [PATCH v22 03/13] power: reset: reboot-mode: Add support for predefined reboot modes Shivendra Pratap
2026-05-14 21:09   ` sashiko-bot
2026-05-14 14:25 ` [PATCH v22 04/13] firmware: psci: Introduce command-based resets Shivendra Pratap
2026-05-14 21:23   ` sashiko-bot
2026-05-14 14:25 ` [PATCH v22 05/13] mfd: psci-mfd: Add PSCI MFD driver for cpuidle-psci-domain cell Shivendra Pratap
2026-05-14 14:25 ` [PATCH v22 06/13] dt-bindings: arm: Document reboot mode magic Shivendra Pratap
2026-05-14 14:25 ` [PATCH v22 07/13] power: reset: Add psci-reboot-mode driver Shivendra Pratap
2026-05-14 22:49   ` sashiko-bot
2026-05-14 14:25 ` [PATCH v22 08/13] mfd: core: Add firmware-node support to MFD cells Shivendra Pratap
2026-05-14 23:19   ` sashiko-bot [this message]
2026-05-14 14:25 ` [PATCH v22 09/13] mfd: psci-mfd: Add psci-reboot-mode child cell Shivendra Pratap
2026-05-14 14:25 ` [PATCH v22 10/13] arm64: dts: qcom: Add psci reboot-modes for kodiak boards Shivendra Pratap
2026-05-14 14:25 ` [PATCH v22 11/13] arm64: dts: qcom: Add psci reboot-modes for lemans boards Shivendra Pratap
2026-05-14 14:25 ` [PATCH v22 12/13] arm64: dts: qcom: Add psci reboot-modes for monaco boards Shivendra Pratap
2026-05-15  0:24   ` sashiko-bot
2026-05-14 14:25 ` [PATCH v22 13/13] arm64: dts: qcom: Add psci reboot-modes for talos boards Shivendra Pratap

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=20260514231943.93B30C2BCB3@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=krzk+dt@kernel.org \
    --cc=robh@kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=shivendra.pratap@oss.qualcomm.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