[PATCH net] net: hns: fix return value check in __lb_other_process()

[PATCH net] net: hns: fix return value check in __lb_other_process()

[wangyunjian]

From: Yunjian Wang <wangyunjian@huawei.com>

The function skb_copy() could return NULL, the return value
need to be checked.

Fixes: b5996f11ea54 ("net: add Hisilicon Network Subsystem basic ethernet support")
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
---
 drivers/net/ethernet/hisilicon/hns/hns_ethtool.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
index 7165da0ee9aa..ad18f0e20a23 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
@@ -415,6 +415,10 @@ static void __lb_other_process(struct hns_nic_ring_data *ring_data,
 	/* for mutl buffer*/
 	new_skb = skb_copy(skb, GFP_ATOMIC);
 	dev_kfree_skb_any(skb);
+	if (!new_skb) {
+		ndev->stats.rx_dropped++;
+		return;
+	}
 	skb = new_skb;
 
 	check_ok = 0;
-- 
2.23.0

Re: [PATCH net] net: hns: fix return value check in __lb_other_process()

[Yonglong Liu]

On 2020/12/25 18:35, wangyunjian wrote:
> From: Yunjian Wang <wangyunjian@huawei.com>
>
> The function skb_copy() could return NULL, the return value
> need to be checked.
>
> Fixes: b5996f11ea54 ("net: add Hisilicon Network Subsystem basic ethernet support")
> Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
> ---
>   drivers/net/ethernet/hisilicon/hns/hns_ethtool.c | 4 ++++
>   1 file changed, 4 insertions(+)
>
> diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
> index 7165da0ee9aa..ad18f0e20a23 100644
> --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
> +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
> @@ -415,6 +415,10 @@ static void __lb_other_process(struct hns_nic_ring_data *ring_data,
>   	/* for mutl buffer*/
>   	new_skb = skb_copy(skb, GFP_ATOMIC);
>   	dev_kfree_skb_any(skb);
> +	if (!new_skb) {

> +		ndev->stats.rx_dropped++;
You can add a new drop type to ring->states, and then add it to 
ndev->stats.rx_dropped,

then you can know that the rx_dropped is from skb_copy() failed.

This process is from self-test, maybe you can just add a error message 
rather than increase

the ndev->stats.rx_dropped.


> +		return;
> +	}
>   	skb = new_skb;
>   
>   	check_ok = 0;

RE: [PATCH net] net: hns: fix return value check in __lb_other_process()

[wangyunjian]

> -----Original Message-----
> From: liuyonglong
> Sent: Saturday, December 26, 2020 2:39 PM
> To: wangyunjian <wangyunjian@huawei.com>; netdev@vger.kernel.org
> Cc: Zhuangyuzeng (Yisen) <yisen.zhuang@huawei.com>; Salil Mehta
> <salil.mehta@huawei.com>; Lilijun (Jerry) <jerry.lilijun@huawei.com>;
> xudingke <xudingke@huawei.com>
> Subject: Re: [PATCH net] net: hns: fix return value check in
> __lb_other_process()
> 
> 
> On 2020/12/25 18:35, wangyunjian wrote:
> > From: Yunjian Wang <wangyunjian@huawei.com>
> >
> > The function skb_copy() could return NULL, the return value need to be
> > checked.
> >
> > Fixes: b5996f11ea54 ("net: add Hisilicon Network Subsystem basic
> > ethernet support")
> > Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
> > ---
> >   drivers/net/ethernet/hisilicon/hns/hns_ethtool.c | 4 ++++
> >   1 file changed, 4 insertions(+)
> >
> > diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
> > b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
> > index 7165da0ee9aa..ad18f0e20a23 100644
> > --- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
> > +++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
> > @@ -415,6 +415,10 @@ static void __lb_other_process(struct
> hns_nic_ring_data *ring_data,
> >   	/* for mutl buffer*/
> >   	new_skb = skb_copy(skb, GFP_ATOMIC);
> >   	dev_kfree_skb_any(skb);
> > +	if (!new_skb) {
> 
> > +		ndev->stats.rx_dropped++;
> You can add a new drop type to ring->states, and then add it to
> ndev->stats.rx_dropped,
> 
> then you can know that the rx_dropped is from skb_copy() failed.
> 
> This process is from self-test, maybe you can just add a error message rather
> than increase

Thanks for your suggestion, I will update it in next version.

> 
> the ndev->stats.rx_dropped.
> 
> 
> > +		return;
> > +	}
> >   	skb = new_skb;
> >
> >   	check_ok = 0;