Re: [PATCH] uacce: fix concurrency of fops_open and uacce_remove

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 





On 2022/6/17 下午4:20, Zhangfei Gao wrote:


On 2022/6/17 下午2:05, Zhangfei Gao wrote:


On 2022/6/16 下午4:14, Jean-Philippe Brucker wrote:
On Thu, Jun 16, 2022 at 12:10:18PM +0800, Zhangfei Gao wrote:
diff --git a/drivers/misc/uacce/uacce.c b/drivers/misc/uacce/uacce.c
index 281c54003edc..b6219c6bfb48 100644
--- a/drivers/misc/uacce/uacce.c
+++ b/drivers/misc/uacce/uacce.c
@@ -136,9 +136,16 @@ static int uacce_fops_open(struct inode *inode, struct file *filep)
       if (!q)
           return -ENOMEM;
+    mutex_lock(&uacce->queues_lock);
+
+    if (!uacce->parent->driver) {
I don't think this is useful, because the core clears parent->driver after
having run uacce_remove():

    rmmod hisi_zip        open()
     ...                 uacce_fops_open()
     __device_release_driver()      ...
      pci_device_remove()
       hisi_zip_remove()
        hisi_qm_uninit()
         uacce_remove()
          ...              ...
                       mutex_lock(uacce->queues_lock)
      ...                  if (!uacce->parent->driver)
      device_unbind_cleanup()      /* driver still valid, proceed */
       dev->driver = NULL
The check  if (!uacce->parent->driver) is required, otherwise NULL pointer
may happen.
I agree we need something, what I mean is that this check is not
sufficient.

iommu_sva_bind_device
const struct iommu_ops *ops = dev_iommu_ops(dev);  ->
dev->iommu->iommu_dev->ops

rmmod has no issue, but remove parent pci device has the issue.
Ah right, relying on the return value of bind() wouldn't be enough even if
we mandated SVA.

[...]
I think we need the global uacce_mutex to serialize uacce_remove() and
uacce_fops_open(). uacce_remove() would do everything, including
xa_erase(), while holding that mutex. And uacce_fops_open() would try to obtain the uacce object from the xarray while holding the mutex, which
fails if the uacce object is being removed.
Since fops_open get char device refcount, uacce_release will not happen
until open returns.
The refcount only ensures that the uacce_device object is not freed as
long as there are open fds. But uacce_remove() can run while there are
open fds, or fds in the process of being opened. And atfer uacce_remove()
runs, the uacce_device object still exists but is mostly unusable. For
example once the module is freed, uacce->ops is not valid anymore. But
currently uacce_fops_open() may dereference the ops in this case:

    uacce_fops_open()
     if (!uacce->parent->driver)
     /* Still valid, keep going */
     ...                    rmmod
                         uacce_remove()
     ...                     free_module()
     uacce->ops->get_queue() /* BUG */

uacce_remove should wait for uacce->queues_lock, until fops_open release the lock. If open happen just after the uacce_remove: unlock, uacce_bind_queue in open should fail.

Accessing uacce->ops after free_module() is a use-after-free. We need all
you men parent release the resources.
the fops to synchronize with uacce_remove() to ensure they don't use any
resource of the parent after it's been freed.
After fops_open, currently we are counting on parent driver stop all dma first, then call uacce_remove, which is assumption. Like drivers/crypto/hisilicon/zip/zip_main.c: hisi_qm_wait_task_finish, which will wait uacce_release.
If comments this , there may other issue,
Unable to handle kernel paging request at virtual address ffff80000b700204
pc : hisi_qm_cache_wb.part.0+0x2c/0xa0

I see uacce_fops_poll() may have the same problem, and should be inside
uacce_mutex.
Do we need consider this, uacce_remove can happen anytime but not waiting dma stop?

Not sure uacce_mutex can do this.
Currently the sequence is
mutex_lock(&uacce->queues_lock);
mutex_lock(&uacce_mutex);

Or we set all the callbacks of uacce_ops to NULL?
How about in uacce_remove
mutex_lock(&uacce_mutex);
uacce->ops = NULL;
mutex_unlock(&uacce_mutex);

And check uacce->ops  first when using.


Diff like this, will merge together.

 drivers/misc/uacce/uacce.c | 65 ++++++++++++++++++++++++++++++++------
 1 file changed, 56 insertions(+), 9 deletions(-)

diff --git a/drivers/misc/uacce/uacce.c b/drivers/misc/uacce/uacce.c
index b6219c6bfb48..311192728132 100644
--- a/drivers/misc/uacce/uacce.c
+++ b/drivers/misc/uacce/uacce.c
@@ -23,6 +23,11 @@ static int uacce_start_queue(struct uacce_queue *q)
         goto out_with_lock;
     }

+    if (!q->uacce->ops) {
+        ret = -EINVAL;
+        goto out_with_lock;
+    }
+
     if (q->uacce->ops->start_queue) {
         ret = q->uacce->ops->start_queue(q);
         if (ret < 0)
@@ -46,6 +51,9 @@ static int uacce_put_queue(struct uacce_queue *q)
     if (q->state == UACCE_Q_ZOMBIE)
         goto out;

+    if (!uacce->ops)
+        goto out;
+
     if ((q->state == UACCE_Q_STARTED) && uacce->ops->stop_queue)
         uacce->ops->stop_queue(q);

@@ -65,6 +73,7 @@ static long uacce_fops_unl_ioctl(struct file *filep,
 {
     struct uacce_queue *q = filep->private_data;
     struct uacce_device *uacce = q->uacce;
+    long ret;

     switch (cmd) {
     case UACCE_CMD_START_Q:
@@ -74,10 +83,17 @@ static long uacce_fops_unl_ioctl(struct file *filep,
         return uacce_put_queue(q);

     default:
-        if (!uacce->ops->ioctl)
-            return -EINVAL;
+        mutex_lock(&uacce_mutex);
+
+        if (!uacce->ops || !uacce->ops->ioctl) {
+            ret = -EINVAL;
+            goto out_with_lock;
+        }

-        return uacce->ops->ioctl(q, cmd, arg);
+        ret = uacce->ops->ioctl(q, cmd, arg);
+out_with_lock:
+        mutex_unlock(&uacce_mutex);
+        return ret;
     }
 }

@@ -138,10 +154,13 @@ static int uacce_fops_open(struct inode *inode, struct file *filep)

     mutex_lock(&uacce->queues_lock);

-    if (!uacce->parent->driver) {
+    mutex_lock(&uacce_mutex);
+    if (!uacce->parent || !uacce->ops) {
+        mutex_unlock(&uacce_mutex);
         ret = -ENODEV;
         goto out_with_lock;
     }
+    mutex_unlock(&uacce_mutex);

     ret = uacce_bind_queue(uacce, q);
     if (ret)
@@ -226,6 +245,11 @@ static int uacce_fops_mmap(struct file *filep, struct vm_area_struct *vma)

     mutex_lock(&uacce_mutex);

+    if (!uacce->ops) {
+        ret = -EINVAL;
+        goto out_with_lock;
+    }
+
     if (q->state != UACCE_Q_INIT && q->state != UACCE_Q_STARTED) {
         ret = -EINVAL;
         goto out_with_lock;
@@ -271,9 +295,18 @@ static __poll_t uacce_fops_poll(struct file *file, poll_table *wait)
     struct uacce_device *uacce = q->uacce;

     poll_wait(file, &q->wait, wait);
-    if (uacce->ops->is_q_updated && uacce->ops->is_q_updated(q))
+
+    mutex_lock(&uacce_mutex);
+    if (!uacce->ops)
+        goto out_with_lock;
+
+    if (uacce->ops->is_q_updated && uacce->ops->is_q_updated(q)) {
+        mutex_unlock(&uacce_mutex);
         return EPOLLIN | EPOLLRDNORM;
+    }

+out_with_lock:
+    mutex_unlock(&uacce_mutex);
     return 0;
 }

@@ -312,12 +345,20 @@ static ssize_t available_instances_show(struct device *dev,
                     char *buf)
 {
     struct uacce_device *uacce = to_uacce_device(dev);
+    ssize_t ret;

-    if (!uacce->ops->get_available_instances)
-        return -ENODEV;
+    mutex_lock(&uacce_mutex);
+    if (!uacce->ops || !uacce->ops->get_available_instances) {
+        ret = -ENODEV;
+        goto out_with_lock;
+    }
+
+    ret = sysfs_emit(buf, "%d\n",
+             uacce->ops->get_available_instances(uacce));

-    return sysfs_emit(buf, "%d\n",
-               uacce->ops->get_available_instances(uacce));
+out_with_lock:
+    mutex_unlock(&uacce_mutex);
+    return ret;
 }

 static ssize_t algorithms_show(struct device *dev,
@@ -523,6 +564,12 @@ void uacce_remove(struct uacce_device *uacce)

     /* disable sva now since no opened queues */
     uacce_disable_sva(uacce);
+
+    mutex_lock(&uacce_mutex);
+    uacce->parent = NULL;
+    uacce->ops = NULL;
+    mutex_unlock(&uacce_mutex);
+
     mutex_unlock(&uacce->queues_lock);

     if (uacce->cdev)
--
2.25.1




[Index of Archives]     [Kernel]     [Gnu Classpath]     [Gnu Crypto]     [DM Crypt]     [Netfilter]     [Bugtraq]
  Powered by Linux