The buffer skb is freed via dev_kfree_skb in a loop. After freeing skb,
the value of packet_count is updated via packet_count++. If packet_count
happens to equal the upper bound (i.e., budget), the loop will be broken
and skb may be assigned to desc_data->state.skb. Resulting that
desc_data->state.skb may point to a freed memory chunk. To fix this, the
patch sets skb to NULL after dev_kfree_skb(skb).
Signed-off-by: Pan Bian <[email protected]>
---
drivers/net/ethernet/synopsys/dwc-xlgmac-net.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c b/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c
index 1f8e960..a0d28c4 100644
--- a/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c
+++ b/drivers/net/ethernet/synopsys/dwc-xlgmac-net.c
@@ -1180,6 +1180,7 @@ static int xlgmac_rx_poll(struct xlgmac_channel *channel, int budget)
netif_err(pdata, rx_err, netdev,
"error in received packet\n");
dev_kfree_skb(skb);
+ skb = NULL;
goto next_packet;
}
@@ -1227,6 +1228,7 @@ static int xlgmac_rx_poll(struct xlgmac_channel *channel, int budget)
netif_err(pdata, rx_err, netdev,
"packet length exceeds configured MTU\n");
dev_kfree_skb(skb);
+ skb = NULL;
goto next_packet;
}
--
2.7.4
On 28-11-2018 09:20, Pan Bian wrote:
> The buffer skb is freed via dev_kfree_skb in a loop. After freeing skb,
> the value of packet_count is updated via packet_count++. If packet_count
> happens to equal the upper bound (i.e., budget), the loop will be broken
> and skb may be assigned to desc_data->state.skb. Resulting that
> desc_data->state.skb may point to a freed memory chunk. To fix this, the
> patch sets skb to NULL after dev_kfree_skb(skb).
>
> Signed-off-by: Pan Bian <[email protected]>
This is missing the Fixes tag and your patch prefix should be
[PATCH net].
Thanks and Best Regards,
Jose Miguel Abreu
From: Pan Bian <[email protected]>
Date: Wed, 28 Nov 2018 17:20:53 +0800
> @@ -1180,6 +1180,7 @@ static int xlgmac_rx_poll(struct xlgmac_channel *channel, int budget)
> netif_err(pdata, rx_err, netdev,
> "error in received packet\n");
> dev_kfree_skb(skb);
> + skb = NULL;
> goto next_packet;
> }
>
> @@ -1227,6 +1228,7 @@ static int xlgmac_rx_poll(struct xlgmac_channel *channel, int budget)
> netif_err(pdata, rx_err, netdev,
> "packet length exceeds configured MTU\n");
> dev_kfree_skb(skb);
> + skb = NULL;
> goto next_packet;
> }
Same problem as your amd-xgbe patch, the condition you say can occurr is
actually impossible.