diff mbox

[net,v2,0/2] net: ethoc: Error path and transmit fixes

Message ID 20160712135109.GB19495@octofox.metropolis
State RFC, archived
Delegated to: David Miller
Headers show

Commit Message

Max Filippov July 12, 2016, 1:51 p.m. UTC 
Hello,

On Mon, Jul 11, 2016 at 04:35:53PM -0700, Florian Fainelli wrote:
> This patch series contains two patches for the ethoc driver while testing on a
> TS-7300 board where ethoc is provided by an on-board FPGA.
> 
> First patch was cooked after chasing crashes with invalid resources passed to
> the driver.
> 
> Second patch was cooked after seeing that an interface configured with IP
> 192.168.2.2 was sending ARP packets for 192.168.0.0, no wonder why it could not
> work.

I can see opencores intrerface sending ARP packets shorter than 64 bytes,
but I couldn't reproduce truncation that affects packet contents on my
hardware.

> I don't have access to any other platform using an ethoc interface so
> it could be good to some testing on Xtensa for instance.

I've tested success and error paths affected by this series with the
following additional change on top of it:


Without it the interface becomes non-functional after the first error
in skb_put_padto.

Tested-by: Max Filippov <jcmvbkbc@gmail.com>
Reviewed-by: Max Filippov <jcmvbkbc@gmail.com>

Comments

David Miller July 12, 2016, 5:35 p.m. UTC | #1 
From: Max Filippov <jcmvbkbc@gmail.com>
Date: Tue, 12 Jul 2016 16:51:10 +0300

> Hello,
> 
> On Mon, Jul 11, 2016 at 04:35:53PM -0700, Florian Fainelli wrote:
>> This patch series contains two patches for the ethoc driver while testing on a
>> TS-7300 board where ethoc is provided by an on-board FPGA.
>> 
>> First patch was cooked after chasing crashes with invalid resources passed to
>> the driver.
>> 
>> Second patch was cooked after seeing that an interface configured with IP
>> 192.168.2.2 was sending ARP packets for 192.168.0.0, no wonder why it could not
>> work.
> 
> I can see opencores intrerface sending ARP packets shorter than 64 bytes,
> but I couldn't reproduce truncation that affects packet contents on my
> hardware.
> 
>> I don't have access to any other platform using an ethoc interface so
>> it could be good to some testing on Xtensa for instance.
> 
> I've tested success and error paths affected by this series with the
> following additional change on top of it:
> 
> diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c
> index ca678d4..8c94f45 100644
> --- a/drivers/net/ethernet/ethoc.c
> +++ b/drivers/net/ethernet/ethoc.c
> @@ -862,7 +862,7 @@ static netdev_tx_t ethoc_start_xmit(struct sk_buff *skb, struct net_device *dev)
>  
>  	if (skb_put_padto(skb, ETHOC_ZLEN)) {
>  		dev->stats.tx_errors++;
> -		goto out;
> +		return NETDEV_TX_OK;
>  	}
>  
>  	if (unlikely(skb->len > ETHOC_BUFSIZ)) {
> 
> Without it the interface becomes non-functional after the first error
> in skb_put_padto.
> 
> Tested-by: Max Filippov <jcmvbkbc@gmail.com>
> Reviewed-by: Max Filippov <jcmvbkbc@gmail.com>

Indeed, skb_put_padto() frees the skb on error so this would have caused
a double free.

Florian, please respin this series with the fix and tags added.

Thanks.
diff mbox

Patch

diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c
index ca678d4..8c94f45 100644
--- a/drivers/net/ethernet/ethoc.c
+++ b/drivers/net/ethernet/ethoc.c
@@ -862,7 +862,7 @@  static netdev_tx_t ethoc_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (skb_put_padto(skb, ETHOC_ZLEN)) {
 		dev->stats.tx_errors++;
-		goto out;
+		return NETDEV_TX_OK;
 	}
 
 	if (unlikely(skb->len > ETHOC_BUFSIZ)) {