From: "Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>
To: Marco Scardovi <scardracs@disroot.org>
Cc: corentin.chary@gmail.com, denis.benato@linux.dev,
Hans de Goede <hansg@kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
luke@ljones.dev, platform-driver-x86@vger.kernel.org
Subject: Re: [PATCH v2 1/1] platform/x86: asus-armoury: fix Use-After-Free and memory leak in driver init
Date: Fri, 3 Jul 2026 12:59:40 +0300 (EEST) [thread overview]
Message-ID: <d0f5dc85-32cf-c5bb-e196-814bc1e7a22f@linux.intel.com> (raw)
In-Reply-To: <20260701164333.5219-2-scardracs@disroot.org>
On Wed, 1 Jul 2026, Marco Scardovi wrote:
> In init_rog_tunables(), if dc_limits are defined and allocating
> dc_rog_tunables fails, the already allocated ac_rog_tunables gets
> freed but the pointer stored in
> asus_armoury.rog_tunables[ASUS_ROG_TUNABLE_AC] is not cleared. Since
> init_rog_tunables() returns void, the driver initialization continues,
> which can lead to a Use-After-Free (UAF) when asus_fw_attr_add()
> accesses the freed AC tunables pointer.
>
> Additionally, if init_rog_tunables() succeeds but asus_fw_attr_add()
> fails, the allocated tunables are not freed, resulting in a memory leak.
>
> Fix these issues by making init_rog_tunables() return an error code and
> propagating it in asus_fw_init(). Defer setting the global pointers in
> asus_armoury.rog_tunables until both tunables have been successfully
> allocated. If asus_fw_attr_add() fails, release the allocated resources
> using a standard goto rollback block in asus_fw_init().
>
> Signed-off-by: Marco Scardovi <scardracs@disroot.org>
> ---
> v2:
> - Restructure init_rog_tunables() to use local pointers and defer assignment
> to the global struct until all allocations succeed, eliminating the need
> to set global pointers to NULL on error.
> - Use a goto-rollback pattern in asus_fw_init() for cleaner error handling
> as requested by Ilpo.
>
> drivers/platform/x86/asus-armoury.c | 38 ++++++++++++++++++-----------
> 1 file changed, 24 insertions(+), 14 deletions(-)
>
> diff --git a/drivers/platform/x86/asus-armoury.c b/drivers/platform/x86/asus-armoury.c
> index 495dc1e31d40..ea53a005bb25 100644
> --- a/drivers/platform/x86/asus-armoury.c
> +++ b/drivers/platform/x86/asus-armoury.c
> @@ -989,7 +989,7 @@ static int asus_fw_attr_add(void)
> /* Init / exit ****************************************************************/
>
> /* Set up the min/max and defaults for ROG tunables */
> -static void init_rog_tunables(void)
> +static int init_rog_tunables(void)
> {
> const struct power_limits *ac_limits, *dc_limits;
> struct rog_tunables *ac_rog_tunables = NULL, *dc_rog_tunables = NULL;
> @@ -1000,24 +1000,23 @@ static void init_rog_tunables(void)
> dmi_id = dmi_first_match(power_limits);
> if (!dmi_id) {
> pr_warn("No matching power limits found for this system\n");
> - return;
> + return 0;
> }
>
> /* Get the power data for this system */
> power_data = dmi_id->driver_data;
> if (!power_data) {
> pr_info("No power data available for this system\n");
> - return;
> + return 0;
> }
>
> /* Initialize AC power tunables */
> ac_limits = power_data->ac_data;
> if (ac_limits) {
> - ac_rog_tunables = kzalloc_obj(*asus_armoury.rog_tunables[ASUS_ROG_TUNABLE_AC]);
> + ac_rog_tunables = kzalloc_obj(*ac_rog_tunables);
> if (!ac_rog_tunables)
> - goto err_nomem;
> + return -ENOMEM;
>
> - asus_armoury.rog_tunables[ASUS_ROG_TUNABLE_AC] = ac_rog_tunables;
> ac_rog_tunables->power_limits = ac_limits;
>
> /* Set initial AC values */
> @@ -1060,13 +1059,12 @@ static void init_rog_tunables(void)
> /* Initialize DC power tunables */
> dc_limits = power_data->dc_data;
> if (dc_limits) {
> - dc_rog_tunables = kzalloc_obj(*asus_armoury.rog_tunables[ASUS_ROG_TUNABLE_DC]);
> + dc_rog_tunables = kzalloc_obj(*dc_rog_tunables);
> if (!dc_rog_tunables) {
> kfree(ac_rog_tunables);
> - goto err_nomem;
> + return -ENOMEM;
> }
>
> - asus_armoury.rog_tunables[ASUS_ROG_TUNABLE_DC] = dc_rog_tunables;
> dc_rog_tunables->power_limits = dc_limits;
>
> /* Set initial DC values */
> @@ -1106,15 +1104,16 @@ static void init_rog_tunables(void)
> pr_debug("No DC PPT limits defined\n");
> }
>
> - return;
> + asus_armoury.rog_tunables[ASUS_ROG_TUNABLE_AC] = ac_rog_tunables;
> + asus_armoury.rog_tunables[ASUS_ROG_TUNABLE_DC] = dc_rog_tunables;
>
> -err_nomem:
> - pr_err("Failed to allocate memory for tunables\n");
> + return 0;
This works as well but now we also want to convert it to use __free()
+ no_free_ptr() so we can get rid of that manual kfree() to remove
potential for future mem management errors.
...If you could add a second patch to this series to that effect, it would
great.
--
i.
> }
>
> static int __init asus_fw_init(void)
> {
> char *wmi_uid;
> + int err;
>
> wmi_uid = wmi_get_acpi_device_uid(ASUS_WMI_MGMT_GUID);
> if (!wmi_uid)
> @@ -1127,10 +1126,21 @@ static int __init asus_fw_init(void)
> if (!strcmp(wmi_uid, ASUS_ACPI_UID_ASUSWMI))
> return -ENODEV;
>
> - init_rog_tunables();
> + err = init_rog_tunables();
> + if (err)
> + return err;
>
> /* Must always be last step to ensure data is available */
> - return asus_fw_attr_add();
> + err = asus_fw_attr_add();
> + if (err)
> + goto err_free_tunables;
> +
> + return 0;
> +
> +err_free_tunables:
> + kfree(asus_armoury.rog_tunables[ASUS_ROG_TUNABLE_AC]);
> + kfree(asus_armoury.rog_tunables[ASUS_ROG_TUNABLE_DC]);
> + return err;
> }
>
> static void __exit asus_fw_exit(void)
>
prev parent reply other threads:[~2026-07-03 9:59 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-01 16:42 [PATCH v2 0/1] platform/x86: asus-armoury: fix Use-After-Free and memory leak in driver init Marco Scardovi
2026-07-01 16:42 ` [PATCH v2 1/1] " Marco Scardovi
2026-07-03 9:59 ` Ilpo Järvinen [this message]
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=d0f5dc85-32cf-c5bb-e196-814bc1e7a22f@linux.intel.com \
--to=ilpo.jarvinen@linux.intel.com \
--cc=corentin.chary@gmail.com \
--cc=denis.benato@linux.dev \
--cc=hansg@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luke@ljones.dev \
--cc=platform-driver-x86@vger.kernel.org \
--cc=scardracs@disroot.org \
/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