Linux Input/HID development
 help / color / mirror / Atom feed
From: "Jose Villaseñor Montfort" <pepemontfort@gmail.com>
To: Jiri Kosina <jikos@kernel.org>, Benjamin Tissoires <bentiss@kernel.org>
Cc: linux-input@vger.kernel.org, linux-kernel@vger.kernel.org,
	"Jose Villaseñor Montfort" <pepemontfort@gmail.com>
Subject: [PATCH] HID: magicmouse: prevent unbounded recursion in magicmouse_raw_event()
Date: Tue, 14 Jul 2026 23:35:26 -0600	[thread overview]
Message-ID: <20260715053526.574725-1-pepemontfort@gmail.com> (raw)

magicmouse_raw_event() handles DOUBLE_REPORT_ID (0xf7) packets, which pack
two touch reports into one, by splitting the packet and calling itself on
each half. The only guard against runaway recursion is a "size < 1" check,
which stops zero-sized calls but does not bound the recursion depth.

A malicious HID device that matches this driver can send a report starting
with DOUBLE_REPORT_ID and filled with the sequence [0xf7, 0x00]. Each level
consumes two bytes and recurses on the remainder, so an incoming report of
up to HID_MAX_BUFFER_SIZE (16 KiB) drives roughly 8000 nested calls. That
easily exhausts the 16 KiB kernel stack, leading to a stack overflow: a
panic with CONFIG_VMAP_STACK, or memory corruption without it.

A double report only ever wraps two normal reports; it is never
legitimately nested. Refuse to re-enter the DOUBLE_REPORT_ID case from a
recursive call so the recursion depth is bounded to two, while all valid
packets keep being parsed exactly as before.

Fixes: a462230e16ac ("HID: magicmouse: enable Magic Trackpad support")
Link: https://lore.kernel.org/linux-input/20260706181347.700DB1F00A3F@smtp.kernel.org/
Cc: stable@vger.kernel.org
Signed-off-by: Jose Villaseñor Montfort <pepemontfort@gmail.com>
---
Noticed while reviewing hid-magicmouse.c during the discussion of the
parallel Magic Trackpad USB-C battery work [1]. The recursion issue is
independent of that series and is sent as its own fix.

I also considered refactoring the report parsing into a non-recursive
helper that dispatches the two sub-reports iteratively, which removes the
recursion entirely. That is a larger and more intrusive change; this
minimal guard keeps the diff small and is trivial to backport, so I went
with it. Happy to switch to the refactor if maintainers prefer it.

[1] https://lore.kernel.org/linux-input/20260706175507.47288-1-andfed.net@gmail.com/

 drivers/hid/hid-magicmouse.c | 25 ++++++++++++++++++++-----
 1 file changed, 20 insertions(+), 5 deletions(-)

diff --git a/drivers/hid/hid-magicmouse.c b/drivers/hid/hid-magicmouse.c
index 802a3479e..97562765a 100644
--- a/drivers/hid/hid-magicmouse.c
+++ b/drivers/hid/hid-magicmouse.c
@@ -383,8 +383,8 @@ static void magicmouse_emit_touch(struct magicmouse_sc *msc, int raw_id, u8 *tda
 	}
 }
 
-static int magicmouse_raw_event(struct hid_device *hdev,
-		struct hid_report *report, u8 *data, int size)
+static int __magicmouse_raw_event(struct hid_device *hdev,
+		struct hid_report *report, u8 *data, int size, bool nested)
 {
 	struct magicmouse_sc *msc = hid_get_drvdata(hdev);
 	struct input_dev *input = msc->input;
@@ -495,6 +495,15 @@ static int magicmouse_raw_event(struct hid_device *hdev,
 		 * packet.
 		 */
 
+		/*
+		 * A double report only ever wraps two normal reports, so it is
+		 * never nested. Refuse to recurse a second time; otherwise a
+		 * malicious device could chain DOUBLE_REPORT_ID packets to drive
+		 * unbounded recursion and overflow the kernel stack.
+		 */
+		if (nested)
+			return 0;
+
 		/* Ensure that we have at least 2 elements (report type and size) */
 		if (size < 2)
 			return 0;
@@ -506,9 +515,9 @@ static int magicmouse_raw_event(struct hid_device *hdev,
 			return 0;
 		}
 
-		magicmouse_raw_event(hdev, report, data + 2, data[1]);
-		magicmouse_raw_event(hdev, report, data + 2 + data[1],
-			size - 2 - data[1]);
+		__magicmouse_raw_event(hdev, report, data + 2, data[1], true);
+		__magicmouse_raw_event(hdev, report, data + 2 + data[1],
+			size - 2 - data[1], true);
 		return 0;
 	default:
 		return 0;
@@ -534,6 +543,12 @@ static int magicmouse_raw_event(struct hid_device *hdev,
 	return 1;
 }
 
+static int magicmouse_raw_event(struct hid_device *hdev,
+		struct hid_report *report, u8 *data, int size)
+{
+	return __magicmouse_raw_event(hdev, report, data, size, false);
+}
+
 static int magicmouse_event(struct hid_device *hdev, struct hid_field *field,
 		struct hid_usage *usage, __s32 value)
 {
-- 
2.55.0


             reply	other threads:[~2026-07-15  5:36 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-15  5:35 Jose Villaseñor Montfort [this message]
2026-07-15 23:04 ` [PATCH] HID: magicmouse: prevent unbounded recursion in magicmouse_raw_event() Alec Hall

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=20260715053526.574725-1-pepemontfort@gmail.com \
    --to=pepemontfort@gmail.com \
    --cc=bentiss@kernel.org \
    --cc=jikos@kernel.org \
    --cc=linux-input@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.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