mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Hyunjung Ko <hj351016@gmail.com>
To: Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	linux-kernel@vger.kernel.org, stable@vger.kernel.org,
	Hyunjung Ko <hj351016@gmail.com>
Subject: [PATCH] intel_th: msu: drain the window-switch work on removal
Date: Tue, 10 Nov 2026 19:55:11 +0900	[thread overview]
Message-ID: <20261110105511.2488300-1-hj351016@gmail.com> (raw)

msc->work (msc_work) is scheduled from intel_th_msc_interrupt() when the
next window is locked and the trace has to be stopped from process
context; the handler dereferences msc and calls intel_th_msc_deactivate().

msc is allocated with devm_kzalloc(), so devres frees it after
intel_th_msc_remove() returns. remove() deactivates the trace (so no new
work is queued) but never drains a msc->work that the interrupt has
already queued. That pending work can then run after msc has been freed
and dereference freed memory (msc->buf_mutex) -- a use-after-free.

Cancel the work in intel_th_msc_remove(), after the trace is deactivated
so it cannot be re-queued, before the device is torn down.

Found by static analysis; not runtime-reproduced (needs an Intel Trace Hub
with the MSU in stop-on-full mode and a window-full interrupt racing
removal).

Fixes: 615c164da0eb ("intel_th: msu: Introduce buffer interface")
Cc: stable@vger.kernel.org
Signed-off-by: Hyunjung Ko <hj351016@gmail.com>
---
 drivers/hwtracing/intel_th/msu.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/hwtracing/intel_th/msu.c b/drivers/hwtracing/intel_th/msu.c
index 84d99d7..902e7d2 100644
--- a/drivers/hwtracing/intel_th/msu.c
+++ b/drivers/hwtracing/intel_th/msu.c
@@ -2174,6 +2174,14 @@ static void intel_th_msc_remove(struct intel_th_device *thdev)
 	intel_th_msc_deactivate(thdev);

 	/*
+	 * msc->work (msc_work) is scheduled from the window-switch interrupt
+	 * when stop_on_full is set and dereferences msc; deactivate has stopped
+	 * the trace so no new work is queued, drain a pending one before the
+	 * devm-allocated msc is freed.
+	 */
+	cancel_work_sync(&msc->work);
+
+	/*
 	 * Buffers should not be used at this point except if the
 	 * output character device is still open and the parent
 	 * device gets detached from its bus, which is a FIXME.
--
2.43.0

                 reply	other threads:[~2026-11-10 10:55 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=20261110105511.2488300-1-hj351016@gmail.com \
    --to=hj351016@gmail.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=andriy.shevchenko@linux.intel.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@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

all inboxes | Powered by JetHome®