CVE-2022-44033: [PATCH v3] char: pcmcia: cm4040_cs: Fix use-after-free in reader_fops

From: Hyunwoo Kim [email protected] To: [email protected] Cc: [email protected], [email protected], [email protected], [email protected], [email protected] Subject: [PATCH v3] char: pcmcia: cm4040_cs: Fix use-after-free in reader_fops Date: Sun, 18 Sep 2022 21:04:57 -0700 [thread overview] Message-ID: 20220919040457.GA302681@ubuntu (raw)

A race condition may occur if the user physically removes the pcmcia device while calling open() for this char device node.

This is a race condition between the cm4040_open() function and the reader_detach() function, which may eventually result in UAF.

So, add a refcount check to reader_detach() to free the “dev” structure after the char device node is close()d.

Signed-off-by: Hyunwoo Kim [email protected]

drivers/char/pcmcia/cm4040_cs.c | 50 ++++++++++++++++++++++±--------- 1 file changed, 35 insertions(+), 15 deletions(-)

diff --git a/drivers/char/pcmcia/cm4040_cs.c b/drivers/char/pcmcia/cm4040_cs.c index 827711911da4…bb9116d81890 100644 — a/drivers/char/pcmcia/cm4040_cs.c +++ b/drivers/char/pcmcia/cm4040_cs.c @@ -59,6 +59,7 @@ static DEFINE_MUTEX(cm4040_mutex); /* how often to poll for fifo status change */ #define POLL_PERIOD msecs_to_jiffies(10)

+static void cm4040_delete(struct kref *kref); static void reader_release(struct pcmcia_device *link);

static int major; @@ -73,6 +74,7 @@ struct reader_dev { wait_queue_head_t poll_wait; wait_queue_head_t read_wait; wait_queue_head_t write_wait;

• struct kref refcnt; unsigned long buffer_status; unsigned long timeout; unsigned char s_buf[READ_WRITE_BUFFER_SIZE]; @@ -102,6 +104,28 @@ static inline unsigned char xinb(unsigned short port) } #endif

+static void cm4040_delete(struct kref *kref) +{

• struct reader_dev *dev = container_of(kref, struct reader_dev, refcnt);
• struct pcmcia_device *link = dev->p_dev;
• int devno;
• /* find device */
• for (devno = 0; devno < CM_MAX_DEV; devno++) {

•   if (dev\_table\[devno\] == link)
  

•       break;
  

• }
• if (devno == CM_MAX_DEV)

•   return;
  

• reader_release(link);
• dev_table[devno] = NULL;
• kfree(dev);
• device_destroy(cmx_class, MKDEV(major, devno)); +}

/* poll the device fifo status register. not to be confused with * the poll syscall. */ static void cm4040_do_poll(struct timer_list *t) @@ -442,6 +466,7 @@ static int cm4040_open(struct inode *inode, struct file *filp) return -ENODEV;

mutex\_lock(&cm4040\_mutex);

• link = dev_table[minor]; if (link == NULL || !pcmcia_dev_present(link)) { ret = -ENODEV; @@ -468,8 +493,11 @@ static int cm4040_open(struct inode *inode, struct file *filp)

  DEBUGP(2, dev, “<- cm4040_open (successfully)\n”); ret = nonseekable_open(inode, filp);

• kref_get(&dev->refcnt); out: mutex_unlock(&cm4040_mutex);

• return ret; }

@@ -495,6 +523,9 @@ static int cm4040_close(struct inode *inode, struct file *filp) wake_up(&dev->devq);

DEBUGP(2, dev, "<- cm4040\_close\\n");

• kref_put(&dev->refcnt, cm4040_delete);
• return 0; }

@@ -584,6 +615,7 @@ static int reader_probe(struct pcmcia_device *link) init_waitqueue_head(&dev->read_wait); init_waitqueue_head(&dev->write_wait); timer_setup(&dev->poll_timer, cm4040_do_poll, 0);

• kref_init(&dev->refcnt);

  ret = reader_config(link, i); if (ret) { @@ -600,22 +632,10 @@ static int reader_probe(struct pcmcia_device *link) static void reader_detach(struct pcmcia_device *link) { struct reader_dev *dev = link->priv; - int devno;

• /* find device */
• for (devno = 0; devno < CM_MAX_DEV; devno++) {

•   if (dev\_table\[devno\] == link)
  

•       break;
  

• }
• if (devno == CM_MAX_DEV)

•   return;
  

• reader_release(link);
• dev_table[devno] = NULL;
• kfree(dev);

- device_destroy(cmx_class, MKDEV(major, devno));

• mutex_lock(&cm4040_mutex);

• kref_put(&dev->refcnt, cm4040_delete);

• mutex_unlock(&cm4040_mutex);

  return; } – 2.25.1

Dear,

I fixed the wrong patch referencing “dev” after kref_put() in the previous version of the patch.

Regards, Hyunwoo Kim.

             reply  other threads:\[~2022-09-19  4:05 UTC|newest\]

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=20220919040457.GA302681@ubuntu \ –[email protected] \ –[email protected] \ –[email protected] \ –[email protected] \ –[email protected] \ –[email protected] \ /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; as well as URLs for NNTP newsgroup(s).