Skip to contentSkip to conversation

usbnet:fix NPE during rx_complete

PATCH Applied Ying Lu 2 participants 3 messages Apr 2, 2025 – Apr 4, 2025

Subsystems (6)
lore
Share & tools
lore

Ying Lu

author
#0 lore raw

v4 cover usbnet:fix NPE during rx_complete

From: Ying Lu <[email protected]>

The patchset fix the issue caused by the following modifications:
commit 04e906839a053f092ef53f4fb2d610983412b904 (usbnet: fix cyclical race on disconnect with work queue)

The issue:
The usb_submit_urb function lacks a usbnet_going_away validation, whereas __usbnet_queue_skb includes this check. This inconsistency creates a race condition where: A URB request may succeed, but the corresponding SKB data fails to be queued.

Subsequent processes (e.g., rx_complete → defer_bh → __skb_unlink(skb, list)) attempt to access skb->next, triggering a NULL pointer dereference (Kernel Panic).

Fix issue:
adding the usbnet_going_away check in usb_submit_urb to synchronize the validation logic.

Changes in v4
-use the correct "Cc:" tag format.

Changes in v3
-use the correct "Fixes:" tag format.

Changes in v2
-Use the formal name instead of an email alias.

Ying Lu (1):
usbnet:fix NPE during rx_complete

 drivers/net/usb/usbnet.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
Signature · 1 line
2.49.0
author
#1 lore raw

PATCH v4 1/1 usbnet:fix NPE during rx_complete

From: Ying Lu <[email protected]>

Missing usbnet_going_away Check in Critical Path. The usb_submit_urb function lacks a usbnet_going_away validation, whereas __usbnet_queue_skb includes this check.

This inconsistency creates a race condition where:
A URB request may succeed, but the corresponding SKB data fails to be queued.

Subsequent processes:
(e.g., rx_complete → defer_bh → __skb_unlink(skb, list)) attempt to access skb->next, triggering a NULL pointer dereference (Kernel Panic).

Fixes 04e906839a05 ("usbnet: fix cyclical race on disconnect with work queue")
---
 drivers/net/usb/usbnet.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/usb/usbnet.c +3 -3
index 44179f4e807f..5161bb5d824b 100644
@@ -519,7 +519,8 @@ static int rx_submit (struct usbnet *dev, struct urb *urb, gfp_t flags)
 	    netif_device_present (dev->net) &&
 	    test_bit(EVENT_DEV_OPEN, &dev->flags) &&
 	    !test_bit (EVENT_RX_HALT, &dev->flags) &&
-	    !test_bit (EVENT_DEV_ASLEEP, &dev->flags)) {
+	    !test_bit (EVENT_DEV_ASLEEP, &dev->flags) &&
+	    !usbnet_going_away(dev)) {
 		switch (retval = usb_submit_urb (urb, GFP_ATOMIC)) {
 		case -EPIPE:
 			usbnet_defer_kevent (dev, EVENT_RX_HALT);
@@ -540,8 +541,7 @@ static int rx_submit (struct usbnet *dev, struct urb *urb, gfp_t flags)
 			tasklet_schedule (&dev->bh);
 			break;
 		case 0:
-			if (!usbnet_going_away(dev))
-				__usbnet_queue_skb(&dev->rxq, skb, rx_start);
+			__usbnet_queue_skb(&dev->rxq, skb, rx_start);
 		}
 	} else {
 		netif_dbg(dev, ifdown, dev->net, "rx: stopped\n");
Signature · 1 line
2.49.0
#2 lore raw
In reply to Ying Lu
This patch was applied to netdev/net.git (main) by Jakub Kicinski <[email protected]>:

Hello:

This patch was applied to netdev/net.git (main)
by Jakub Kicinski <[email protected]>:

Quoted 6 lines · On Wed, 2 Apr 2025 16:58:58 +0800 you wrote:

From: Ying Lu <[email protected]>

The patchset fix the issue caused by the following modifications:
commit 04e906839a053f092ef53f4fb2d610983412b904 (usbnet: fix cyclical race on disconnect with work queue)

Quoted 7 lines

The issue:
The usb_submit_urb function lacks a usbnet_going_away validation, whereas __usbnet_queue_skb includes this check. This inconsistency creates a race condition where: A URB request may succeed, but the corresponding SKB data fails to be queued.

[...]

Here is the summary with links:
- [v4,1/1] usbnet:fix NPE during rx_complete

    https://git.kernel.org/netdev/net/c/51de36000934

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html