mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: hyunjungg <hj351016@gmail.com>
To: Lee Jones <lee@kernel.org>
Cc: mfd@lists.linux.dev, linux-kernel@vger.kernel.org,
	Hyunjung Ko <hj351016@gmail.com>
Subject: [PATCH] mfd: si476x-i2c: cancel RDS drainer work before freeing RDS FIFO
Date: Fri,  6 Nov 2026 01:27:22 +0900	[thread overview]
Message-ID: <20261105162722.1688960-1-hj351016@gmail.com> (raw)

From: Hyunjung Ko <hj351016@gmail.com>

The RDS FIFO drainer work (core->rds_fifo_drainer) is scheduled from the
interrupt dispatcher si476x_core_get_and_signal_status() whenever an
SI476X_FM_RDS_INT interrupt is seen:

	if (status & SI476X_FM_RDS_INT)
		si476x_core_start_rds_drainer_once(core);

si476x_core_start_rds_drainer_once() runs the work on the shared system
workqueue via schedule_work(&core->rds_fifo_drainer). The work handler
si476x_core_drain_rds_fifo() dereferences @core and reads from
core->rds_fifo.

si476x_core_remove() does not cancel this work. It disables the interrupt
(or, in polling mode, cancels only the sibling core->status_monitor
delayed work) and then frees the RDS FIFO:

	if (client->irq)
		disable_irq(client->irq);
	else
		cancel_delayed_work_sync(&core->status_monitor);

	kfifo_free(&core->rds_fifo);

disable_irq() waits for the interrupt handler to finish but does not flush
or cancel work that the handler has already queued. A drainer work item
that was scheduled just before (or is still running at) remove time can
therefore run after kfifo_free() and after the devm-allocated @core itself
is freed, dereferencing freed memory and operating on a freed kfifo
(use-after-free). The sibling status_monitor work is drained on teardown
but this one is not.

Drain the RDS drainer work before freeing the FIFO (and before the
devm-managed @core is released) to close the window.

Found by code review; not runtime-reproduced (requires si476x AM/FM radio
hardware and an RDS interrupt racing device removal).

Signed-off-by: Hyunjung Ko <hj351016@gmail.com>
---
 drivers/mfd/si476x-i2c.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/mfd/si476x-i2c.c b/drivers/mfd/si476x-i2c.c
index 0000000..1111111 100644
--- a/drivers/mfd/si476x-i2c.c
+++ b/drivers/mfd/si476x-i2c.c
@@ -831,6 +831,8 @@ static void si476x_core_remove(struct i2c_client *client)
 	if (client->irq)
 		disable_irq(client->irq);
 	else
 		cancel_delayed_work_sync(&core->status_monitor);

+	cancel_work_sync(&core->rds_fifo_drainer);
+
 	kfifo_free(&core->rds_fifo);
 }


                 reply	other threads:[~2026-11-05 16:27 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20261105162722.1688960-1-hj351016@gmail.com \
    --to=hj351016@gmail.com \
    --cc=lee@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mfd@lists.linux.dev \
    /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

all inboxes | Powered by JetHome®