All of lore.kernel.org
 help / color / mirror / Atom feed
From: Shuah Khan <skhan@linuxfoundation.org>
To: valentina.manea.m@gmail.com, shuah@kernel.org,
	gregkh@linuxfoundation.org
Cc: Shuah Khan <skhan@linuxfoundation.org>,
	linux-usb@vger.kernel.org, linux-kernel@vger.kernel.org,
	syzbot+a93fba6d384346a761e3@syzkaller.appspotmail.com,
	stable@vger.kernel.org
Subject: [PATCH 2/4] usbip: stub-dev synchronize sysfs code paths
Date: Mon, 29 Mar 2021 19:36:49 -0600	[thread overview]
Message-ID: <2b182f3561b4a065bf3bf6dce3b0e9944ba17b3f.1616807117.git.skhan@linuxfoundation.org> (raw)
In-Reply-To: <cover.1616807117.git.skhan@linuxfoundation.org>

Fuzzing uncovered race condition between sysfs code paths in usbip
drivers. Device connect/disconnect code paths initiated through
sysfs interface are prone to races if disconnect happens during
connect and vice versa.

Use sysfs_lock to protect sysfs paths in stub-dev.

Reported-and-tested-by: syzbot+a93fba6d384346a761e3@syzkaller.appspotmail.com
Cc: stable@vger.kernel.org
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
---
 drivers/usb/usbip/stub_dev.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/usbip/stub_dev.c b/drivers/usb/usbip/stub_dev.c
index 8f1de1fbbeed..d8d3892e5a69 100644
--- a/drivers/usb/usbip/stub_dev.c
+++ b/drivers/usb/usbip/stub_dev.c
@@ -63,6 +63,7 @@ static ssize_t usbip_sockfd_store(struct device *dev, struct device_attribute *a
 
 		dev_info(dev, "stub up\n");
 
+		mutex_lock(&sdev->ud.sysfs_lock);
 		spin_lock_irq(&sdev->ud.lock);
 
 		if (sdev->ud.status != SDEV_ST_AVAILABLE) {
@@ -87,13 +88,13 @@ static ssize_t usbip_sockfd_store(struct device *dev, struct device_attribute *a
 		tcp_rx = kthread_create(stub_rx_loop, &sdev->ud, "stub_rx");
 		if (IS_ERR(tcp_rx)) {
 			sockfd_put(socket);
-			return -EINVAL;
+			goto unlock_mutex;
 		}
 		tcp_tx = kthread_create(stub_tx_loop, &sdev->ud, "stub_tx");
 		if (IS_ERR(tcp_tx)) {
 			kthread_stop(tcp_rx);
 			sockfd_put(socket);
-			return -EINVAL;
+			goto unlock_mutex;
 		}
 
 		/* get task structs now */
@@ -112,6 +113,8 @@ static ssize_t usbip_sockfd_store(struct device *dev, struct device_attribute *a
 		wake_up_process(sdev->ud.tcp_rx);
 		wake_up_process(sdev->ud.tcp_tx);
 
+		mutex_unlock(&sdev->ud.sysfs_lock);
+
 	} else {
 		dev_info(dev, "stub down\n");
 
@@ -122,6 +125,7 @@ static ssize_t usbip_sockfd_store(struct device *dev, struct device_attribute *a
 		spin_unlock_irq(&sdev->ud.lock);
 
 		usbip_event_add(&sdev->ud, SDEV_EVENT_DOWN);
+		mutex_unlock(&sdev->ud.sysfs_lock);
 	}
 
 	return count;
@@ -130,6 +134,8 @@ static ssize_t usbip_sockfd_store(struct device *dev, struct device_attribute *a
 	sockfd_put(socket);
 err:
 	spin_unlock_irq(&sdev->ud.lock);
+unlock_mutex:
+	mutex_unlock(&sdev->ud.sysfs_lock);
 	return -EINVAL;
 }
 static DEVICE_ATTR_WO(usbip_sockfd);
@@ -270,6 +276,7 @@ static struct stub_device *stub_device_alloc(struct usb_device *udev)
 	sdev->ud.side		= USBIP_STUB;
 	sdev->ud.status		= SDEV_ST_AVAILABLE;
 	spin_lock_init(&sdev->ud.lock);
+	mutex_init(&sdev->ud.sysfs_lock);
 	sdev->ud.tcp_socket	= NULL;
 	sdev->ud.sockfd		= -1;
 
-- 
2.27.0


  parent reply	other threads:[~2021-03-30  1:37 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-30  1:36 [PATCH 0/4] usbip synchronize sysfs code paths Shuah Khan
2021-03-30  1:36 ` [PATCH 1/4] usbip: add sysfs_lock to " Shuah Khan
2021-03-30  1:36 ` Shuah Khan [this message]
2021-03-30  1:36 ` [PATCH 3/4] usbip: vudc " Shuah Khan
2021-03-30  1:36 ` [PATCH 4/4] usbip: synchronize event handler with " Shuah Khan
2021-04-16 20:53 [PATCH 1/4] usbip: add sysfs_lock to synchronize " Tom Seewald
2021-04-16 20:53 ` [PATCH 2/4] usbip: stub-dev " Tom Seewald
2021-04-16 21:44   ` Shuah Khan

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=2b182f3561b4a065bf3bf6dce3b0e9944ba17b3f.1616807117.git.skhan@linuxfoundation.org \
    --to=skhan@linuxfoundation.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=shuah@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=syzbot+a93fba6d384346a761e3@syzkaller.appspotmail.com \
    --cc=valentina.manea.m@gmail.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.