From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 7F782364935; Fri, 7 Aug 2026 14:42:26 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1786113765; cv=none; b=RnvLoPE3J3ooZmSuyVehOH/F9eY/HP61HWMdxXbpZQ+TVpzHSq9De0nFKgHqF+TmSSywpH0qWf5BWN2enH//jk5/ZiA+fsf0pmqUAm+5Zq5fmzv5hSx6bsh49sgJnRBVHZKfYgdDfSK9i/hTq3qkNh+5Ez0GZU8SVmohafknEfI= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1786113765; c=relaxed/simple; bh=ZljBuLB6xpzNNh6pY/bgtJA0VO8yj3Cdgsv9LWTcPyc=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=qG7xRZnTj2KzShWmsB5oQdu7jdnM7VZE5khWl/1FMUQMfizEoOgZIgjRjVbYcjZ8JUK7Rbn8DI4hs2NiY1dC++UttvyfMvwYgDxVjEhdup/baAFKB3I8Q07P1E4Eh5Sudql92hxH59zn9W0PApbwYUH/WXZRXFTHD88tBJkd7rM= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b=1HlL+iJJ; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="1HlL+iJJ" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 408521F00ACF; Fri, 7 Aug 2026 14:42:19 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linuxfoundation.org; s=korg; t=1786113739; bh=+bgTW2IZy+K/hxWrQafDJpCAnomdSqsklAs6fqWsSKM=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=1HlL+iJJ39aqoJEGEl+GPQDBACDPMSbsWMbNAeorfeM41SGBsl5+6tuiLw0jsSgc5 RbbUx08bpaIBpp6QbrxDrMlkBLS+LCCovymWWU/6kkvvsjPjWWOYAHuKE8Lovgfzfe pWetK8206ZYN6RVezdfTQYXGcMiSio5vKz2ynozw= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Michael Bommarito , Mika Westerberg , Sasha Levin Subject: [PATCH 6.12 019/337] thunderbolt: Prevent XDomain delayed work use-after-free on disconnect Date: Fri, 7 Aug 2026 16:33:42 +0200 Message-ID: <20260807143418.935740605@linuxfoundation.org> X-Mailer: git-send-email 2.55.0 In-Reply-To: <20260807143418.516897842@linuxfoundation.org> References: <20260807143418.516897842@linuxfoundation.org> User-Agent: quilt/0.69 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 6.12-stable review patch. If anyone has any objections, please let me know. ------------------ From: Michael Bommarito [ Upstream commit 2c5d2d3c3f70cde2565d7b279b544893a2035842 ] tb_xdp_handle_request() runs on system_wq and queues xd->state_work via queue_delayed_work() in three request handlers: PROPERTIES_CHANGED_REQUEST, UUID_REQUEST (via start_handshake), and LINK_STATE_CHANGE_REQUEST. Similarly, update_xdomain() queues xd->properties_changed_work when local properties change. Concurrently, tb_xdomain_remove() calls stop_handshake() which does cancel_delayed_work_sync() on both delayed works. Later, tb_xdomain_unregister() calls device_unregister() which eventually frees the xdomain. Since commit 559c1e1e0134 ("thunderbolt: Run tb_xdp_handle_request() in system workqueue") moved the request handler off tb->wq, the handler and the remove path are no longer serialized. If queue_delayed_work() executes after cancel_delayed_work_sync() but before the xdomain is freed, the delayed work fires on a freed object. Add xd->removing that tb_xdomain_remove() sets under xd->lock before calling stop_handshake(). Each external queue site holds the same lock and checks removing before calling queue_delayed_work(). This provides the mutual exclusion needed: either the queue site acquires the lock first and queues work that the subsequent cancel will see, or the remove path acquires the lock first and the queue site observes removing == true and skips the queue. Fixes: 559c1e1e0134 ("thunderbolt: Run tb_xdp_handle_request() in system workqueue") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Signed-off-by: Mika Westerberg Signed-off-by: Sasha Levin --- drivers/thunderbolt/xdomain.c | 40 ++++++++++++++++++++++++++--------- include/linux/thunderbolt.h | 3 +++ 2 files changed, 33 insertions(+), 10 deletions(-) diff --git a/drivers/thunderbolt/xdomain.c b/drivers/thunderbolt/xdomain.c index 06357073f4ab8..ffe7ce0737491 100644 --- a/drivers/thunderbolt/xdomain.c +++ b/drivers/thunderbolt/xdomain.c @@ -783,9 +783,13 @@ static void tb_xdp_handle_request(struct work_struct *work) * the xdomain related to this connection as well in * case there is a change in services it offers. */ - if (xd && device_is_registered(&xd->dev)) - queue_delayed_work(tb->wq, &xd->state_work, - msecs_to_jiffies(XDOMAIN_SHORT_TIMEOUT)); + if (xd) { + mutex_lock(&xd->lock); + if (!xd->removing && device_is_registered(&xd->dev)) + queue_delayed_work(tb->wq, &xd->state_work, + msecs_to_jiffies(XDOMAIN_SHORT_TIMEOUT)); + mutex_unlock(&xd->lock); + } break; case UUID_REQUEST_OLD: @@ -798,8 +802,12 @@ static void tb_xdp_handle_request(struct work_struct *work) * received UUID request from the remote host. */ if (!ret && xd && xd->state == XDOMAIN_STATE_ERROR) { - dev_dbg(&xd->dev, "restarting handshake\n"); - start_handshake(xd); + mutex_lock(&xd->lock); + if (!xd->removing) { + dev_dbg(&xd->dev, "restarting handshake\n"); + start_handshake(xd); + } + mutex_unlock(&xd->lock); } break; @@ -827,9 +835,13 @@ static void tb_xdp_handle_request(struct work_struct *work) ret = tb_xdp_link_state_change_response(ctl, route, sequence, 0); - xd->target_link_width = lsc->tlw; - queue_delayed_work(tb->wq, &xd->state_work, - msecs_to_jiffies(XDOMAIN_SHORT_TIMEOUT)); + mutex_lock(&xd->lock); + if (!xd->removing) { + xd->target_link_width = lsc->tlw; + queue_delayed_work(tb->wq, &xd->state_work, + msecs_to_jiffies(XDOMAIN_SHORT_TIMEOUT)); + } + mutex_unlock(&xd->lock); } else { tb_xdp_error_response(ctl, route, sequence, ERROR_NOT_READY); @@ -2068,6 +2080,10 @@ void tb_xdomain_remove(struct tb_xdomain *xd) { tb_xdomain_debugfs_remove(xd); + mutex_lock(&xd->lock); + xd->removing = true; + mutex_unlock(&xd->lock); + stop_handshake(xd); device_for_each_child_reverse(&xd->dev, xd, unregister_service); @@ -2464,8 +2480,12 @@ static int update_xdomain(struct device *dev, void *data) xd = tb_to_xdomain(dev); if (xd) { - queue_delayed_work(xd->tb->wq, &xd->properties_changed_work, - msecs_to_jiffies(50)); + mutex_lock(&xd->lock); + if (!xd->removing) + queue_delayed_work(xd->tb->wq, + &xd->properties_changed_work, + msecs_to_jiffies(50)); + mutex_unlock(&xd->lock); } return 0; diff --git a/include/linux/thunderbolt.h b/include/linux/thunderbolt.h index 7d902d8c054b2..4ef9c43026569 100644 --- a/include/linux/thunderbolt.h +++ b/include/linux/thunderbolt.h @@ -202,6 +202,8 @@ enum tb_link_width { * @link_width: Width of the downstream facing link * @link_usb4: Downstream link is USB4 * @is_unplugged: The XDomain is unplugged + * @removing: Set by tb_xdomain_remove() under @lock to prevent + * concurrent delayed work queueing * @needs_uuid: If the XDomain does not have @remote_uuid it will be * queried first * @service_ids: Used to generate IDs for the services @@ -250,6 +252,7 @@ struct tb_xdomain { enum tb_link_width link_width; bool link_usb4; bool is_unplugged; + bool removing; bool needs_uuid; struct ida service_ids; struct ida in_hopids; -- 2.53.0