From: Kery Qi <qikeyu2017@gmail.com>
To: gregkh@linuxfoundation.org
Cc: linux-usb@vger.kernel.org, Kery Qi <qikeyu2017@gmail.com>
Subject: [PATCH] usb: core: hub: avoid NULL deref in usb_disconnect()
Date: Fri, 9 Jan 2026 06:04:37 +0800 [thread overview]
Message-ID: <20260108220437.1930-1-qikeyu2017@gmail.com> (raw)
usb_disconnect() assumes that usb_hub_to_struct_hub(udev->parent)
always returns a valid struct usb_hub when udev->parent is set.
However, usb_hub_to_struct_hub() can return NULL if the parent hub
is already unconfigured/disconnected (e.g. actconfig == NULL or
maxchild == 0).
In that case, usb_disconnect() would dereference hub->ports[...]
and hub->child_usage_bits, causing a NULL pointer dereference and
kernel crash during device disconnect.
Guard the hub-specific port handling with a NULL check and only
touch port_dev links and the child_usage_bits/pm runtime reference
when a valid hub is present.
Fixes: 7027df36e418 ("usb: resume child device when port is powered on")
Signed-off-by: Kery Qi <qikeyu2017@gmail.com>
---
drivers/usb/core/hub.c | 27 +++++++++++++++------------
1 file changed, 15 insertions(+), 12 deletions(-)
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index be50d03034a9..444e04ea433e 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -2348,19 +2348,21 @@ void usb_disconnect(struct usb_device **pdev)
if (udev->parent) {
port1 = udev->portnum;
hub = usb_hub_to_struct_hub(udev->parent);
- port_dev = hub->ports[port1 - 1];
+ if (hub) {
+ port_dev = hub->ports[port1 - 1];
- sysfs_remove_link(&udev->dev.kobj, "port");
- sysfs_remove_link(&port_dev->dev.kobj, "device");
+ sysfs_remove_link(&udev->dev.kobj, "port");
+ sysfs_remove_link(&port_dev->dev.kobj, "device");
- /*
- * As usb_port_runtime_resume() de-references udev, make
- * sure no resumes occur during removal
- */
- if (!test_and_set_bit(port1, hub->child_usage_bits))
- pm_runtime_get_sync(&port_dev->dev);
+ /*
+ * As usb_port_runtime_resume() de-references udev, make
+ * sure no resumes occur during removal
+ */
+ if (!test_and_set_bit(port1, hub->child_usage_bits))
+ pm_runtime_get_sync(&port_dev->dev);
- typec_deattach(port_dev->connector, &udev->dev);
+ typec_deattach(port_dev->connector, &udev->dev);
+ }
}
usb_remove_ep_devs(&udev->ep0);
@@ -2385,8 +2387,9 @@ void usb_disconnect(struct usb_device **pdev)
*pdev = NULL;
spin_unlock_irq(&device_state_lock);
- if (port_dev && test_and_clear_bit(port1, hub->child_usage_bits))
- pm_runtime_put(&port_dev->dev);
+ if (hub)
+ if (port_dev && test_and_clear_bit(port1, hub->child_usage_bits))
+ pm_runtime_put(&port_dev->dev);
hub_free_dev(udev);
--
2.34.1
next reply other threads:[~2026-01-08 22:04 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-08 22:04 Kery Qi [this message]
2026-01-08 22:23 ` [PATCH] usb: core: hub: avoid NULL deref in usb_disconnect() Alan Stern
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=20260108220437.1930-1-qikeyu2017@gmail.com \
--to=qikeyu2017@gmail.com \
--cc=gregkh@linuxfoundation.org \
--cc=linux-usb@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