Thread (14 messages) read the whole thread 14 messages, 4 authors, 2022-11-15

RE: [PATCH net-next v3] nfc: Allow to create multiple virtual nci devices

From: Bongsu Jeon <bongsu.jeon@samsung.com>
Date: 2022-11-07 02:46:14

On Sat, Nov 5, 2022 at 2:04 AM Dmitry Vyukov[off-list ref] wrote:
The current virtual nci driver is great for testing and fuzzing.
But it allows to create at most one "global" device which does not allow
to run parallel tests and harms fuzzing isolation and reproducibility.
Restructure the driver to allow creation of multiple independent devices.
This should be backwards compatible for existing tests.
I totally agree with you for parallel tests and good design.
Thanks for good idea.
But please check the abnormal situation.
for example virtual device app is closed(virtual_ncidev_close) first and then
virtual nci driver from nci app tries to call virtual_nci_send or virtual_nci_close.
(there would be problem in virtual_nci_send because of already destroyed mutex)
Before this patch, this driver used virtual_ncidev_mode state and nci_mutex that isn't destroyed.
quoted hunk ↗ jump to hunk
Signed-off-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Bongsu Jeon <bongsu.jeon@samsung.com>
Cc: Krzysztof Kozlowski <redacted>
Cc: netdev@vger.kernel.org

---
Changes in v3:
 - free vdev in virtual_ncidev_close()

Changes in v2:
 - check return value of skb_clone()
 - rebase onto currnet net-next
---
 drivers/nfc/virtual_ncidev.c | 147 +++++++++++++++++------------------
 1 file changed, 71 insertions(+), 76 deletions(-)
diff --git a/drivers/nfc/virtual_ncidev.c b/drivers/nfc/virtual_ncidev.c
index 85c06dbb2c449..bb76c7c7cc822 100644
--- a/drivers/nfc/virtual_ncidev.c
+++ b/drivers/nfc/virtual_ncidev.c
@@ -13,12 +13,6 @@
 
 static int virtual_nci_send(struct nci_dev *ndev, struct sk_buff *skb)
 {
-	mutex_lock(&nci_mutex);
-	if (state != virtual_ncidev_enabled) {
-		mutex_unlock(&nci_mutex);
+	struct virtual_nci_dev *vdev = nci_get_drvdata(ndev);
+
+	mutex_lock(&vdev->mtx);
  
  I think this vdev and vdev->mtx are already destroyed so that it would be problem.
+	if (vdev->send_buff) {
+		mutex_unlock(&vdev->mtx);
 		kfree_skb(skb);
-		return 0;
+		return -1;
 	}

	
 static int virtual_ncidev_close(struct inode *inode, struct file *file)
 {
-	mutex_lock(&nci_mutex);
-
-	if (state == virtual_ncidev_enabled) {
-		state = virtual_ncidev_disabling;
-		mutex_unlock(&nci_mutex);
+	struct virtual_nci_dev *vdev = file->private_data;
 
-		nci_unregister_device(ndev);
-		nci_free_device(ndev);
-
-		mutex_lock(&nci_mutex);
-	}
-
-	state = virtual_ncidev_disabled;
-	mutex_unlock(&nci_mutex);
+	nci_unregister_device(vdev->ndev);
+	nci_free_device(vdev->ndev);
+	mutex_destroy(&vdev->mtx);
+	kfree(vdev);
 
 	return 0;
 }
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help