[PATCH net-next,v4] bonding: return -ENOMEM instead of BUG in alb_upper_dev_walk

[PATCH net-next,v4] bonding: return -ENOMEM instead of BUG in alb_upper_dev_walk

[Zhengchao Shao]

If failed to allocate "tags" or could not find the final upper device from
start_dev's upper list in bond_verify_device_path(), only the loopback
detection of the current upper device should be affected, and the system is
no need to be panic.
So return -ENOMEM in alb_upper_dev_walk to stop walking, print some warn
information when failed to allocate memory for vlan tags in
bond_verify_device_path.

I also think that the following function calls
netdev_walk_all_upper_dev_rcu
---->>>alb_upper_dev_walk
---------->>>bond_verify_device_path
From this way, "end device" can eventually be obtained from "start device"
in bond_verify_device_path, IS_ERR(tags) could be instead of
IS_ERR_OR_NULL(tags) in alb_upper_dev_walk.

Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
---
v4: print information instead of warn
v3: return -ENOMEM instead of zero to stop walk
v2: use WARN_ON_ONCE instead of WARN_ON
---
 drivers/net/bonding/bond_alb.c  | 3 ++-
 drivers/net/bonding/bond_main.c | 5 ++++-
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index dc2c7b979656..7edf0fd58c34 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -985,7 +985,8 @@ static int alb_upper_dev_walk(struct net_device *upper,
 	if (netif_is_macvlan(upper) && !strict_match) {
 		tags = bond_verify_device_path(bond->dev, upper, 0);
 		if (IS_ERR_OR_NULL(tags))
-			BUG();
+			return -ENOMEM;
+
 		alb_send_lp_vid(slave, upper->dev_addr,
 				tags[0].vlan_proto, tags[0].vlan_id);
 		kfree(tags);
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 51d47eda1c87..1a40bd08f984 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2967,8 +2967,11 @@ struct bond_vlan_tag *bond_verify_device_path(struct net_device *start_dev,
 
 	if (start_dev == end_dev) {
 		tags = kcalloc(level + 1, sizeof(*tags), GFP_ATOMIC);
-		if (!tags)
+		if (!tags) {
+			net_err_ratelimited("%s: %s: Failed to allocate tags\n",
+					    __func__, start_dev->name);
 			return ERR_PTR(-ENOMEM);
+		}
 		tags[level].vlan_proto = BOND_VLAN_PROTO_NONE;
 		return tags;
 	}
-- 
2.34.1

Re: [PATCH net-next,v4] bonding: return -ENOMEM instead of BUG in alb_upper_dev_walk

[Jay Vosburgh]

Zhengchao Shao <shaozhengchao@huawei.com> wrote:

>If failed to allocate "tags" or could not find the final upper device from
>start_dev's upper list in bond_verify_device_path(), only the loopback
>detection of the current upper device should be affected, and the system is
>no need to be panic.
>So return -ENOMEM in alb_upper_dev_walk to stop walking, print some warn
>information when failed to allocate memory for vlan tags in
>bond_verify_device_path.
>
>I also think that the following function calls
>netdev_walk_all_upper_dev_rcu
>---->>>alb_upper_dev_walk
>---------->>>bond_verify_device_path
>From this way, "end device" can eventually be obtained from "start device"
>in bond_verify_device_path, IS_ERR(tags) could be instead of
>IS_ERR_OR_NULL(tags) in alb_upper_dev_walk.
>
>Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>

Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>

>---
>v4: print information instead of warn
>v3: return -ENOMEM instead of zero to stop walk
>v2: use WARN_ON_ONCE instead of WARN_ON
>---
> drivers/net/bonding/bond_alb.c  | 3 ++-
> drivers/net/bonding/bond_main.c | 5 ++++-
> 2 files changed, 6 insertions(+), 2 deletions(-)
>
>diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
>index dc2c7b979656..7edf0fd58c34 100644
>--- a/drivers/net/bonding/bond_alb.c
>+++ b/drivers/net/bonding/bond_alb.c
>@@ -985,7 +985,8 @@ static int alb_upper_dev_walk(struct net_device *upper,
> 	if (netif_is_macvlan(upper) && !strict_match) {
> 		tags = bond_verify_device_path(bond->dev, upper, 0);
> 		if (IS_ERR_OR_NULL(tags))
>-			BUG();
>+			return -ENOMEM;
>+
> 		alb_send_lp_vid(slave, upper->dev_addr,
> 				tags[0].vlan_proto, tags[0].vlan_id);
> 		kfree(tags);
>diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
>index 51d47eda1c87..1a40bd08f984 100644
>--- a/drivers/net/bonding/bond_main.c
>+++ b/drivers/net/bonding/bond_main.c
>@@ -2967,8 +2967,11 @@ struct bond_vlan_tag *bond_verify_device_path(struct net_device *start_dev,
> 
> 	if (start_dev == end_dev) {
> 		tags = kcalloc(level + 1, sizeof(*tags), GFP_ATOMIC);
>-		if (!tags)
>+		if (!tags) {
>+			net_err_ratelimited("%s: %s: Failed to allocate tags\n",
>+					    __func__, start_dev->name);
> 			return ERR_PTR(-ENOMEM);
>+		}
> 		tags[level].vlan_proto = BOND_VLAN_PROTO_NONE;
> 		return tags;
> 	}
>-- 
>2.34.1
>
>

Re: [PATCH net-next,v4] bonding: return -ENOMEM instead of BUG in alb_upper_dev_walk

[Paolo Abeni]

On Sat, 2023-11-18 at 16:16 +0800, Zhengchao Shao wrote:
> If failed to allocate "tags" or could not find the final upper device from
> start_dev's upper list in bond_verify_device_path(), only the loopback
> detection of the current upper device should be affected, and the system is
> no need to be panic.
> So return -ENOMEM in alb_upper_dev_walk to stop walking, print some warn
> information when failed to allocate memory for vlan tags in
> bond_verify_device_path.
> 
> I also think that the following function calls
> netdev_walk_all_upper_dev_rcu
> ---->>>alb_upper_dev_walk
> ---------->>>bond_verify_device_path
> > From this way, "end device" can eventually be obtained from "start device"
> in bond_verify_device_path, IS_ERR(tags) could be instead of
> IS_ERR_OR_NULL(tags) in alb_upper_dev_walk.
> 
> Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
> ---
> v4: print information instead of warn
> v3: return -ENOMEM instead of zero to stop walk
> v2: use WARN_ON_ONCE instead of WARN_ON
> ---
>  drivers/net/bonding/bond_alb.c  | 3 ++-
>  drivers/net/bonding/bond_main.c | 5 ++++-
>  2 files changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
> index dc2c7b979656..7edf0fd58c34 100644
> --- a/drivers/net/bonding/bond_alb.c
> +++ b/drivers/net/bonding/bond_alb.c
> @@ -985,7 +985,8 @@ static int alb_upper_dev_walk(struct net_device *upper,
>  	if (netif_is_macvlan(upper) && !strict_match) {
>  		tags = bond_verify_device_path(bond->dev, upper, 0);
>  		if (IS_ERR_OR_NULL(tags))
> -			BUG();
> +			return -ENOMEM;
> +
>  		alb_send_lp_vid(slave, upper->dev_addr,
>  				tags[0].vlan_proto, tags[0].vlan_id);
>  		kfree(tags);
> diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
> index 51d47eda1c87..1a40bd08f984 100644
> --- a/drivers/net/bonding/bond_main.c
> +++ b/drivers/net/bonding/bond_main.c
> @@ -2967,8 +2967,11 @@ struct bond_vlan_tag *bond_verify_device_path(struct net_device *start_dev,
>  
>  	if (start_dev == end_dev) {
>  		tags = kcalloc(level + 1, sizeof(*tags), GFP_ATOMIC);
> -		if (!tags)
> +		if (!tags) {
> +			net_err_ratelimited("%s: %s: Failed to allocate tags\n",
> +					    __func__, start_dev->name);

I'm sorry for the repeated back and forth, but checkpatch points out
that the above warning adds little value: if the memory allocation
fails, the mm layer will emit a lot warning comprising the backtrace.

I suggest to drop it, thanks!

Paolo

Re: [PATCH net-next,v4] bonding: return -ENOMEM instead of BUG in alb_upper_dev_walk

[shaozhengchao]

On 2023/11/21 19:41, Paolo Abeni wrote:
> On Sat, 2023-11-18 at 16:16 +0800, Zhengchao Shao wrote:
>> If failed to allocate "tags" or could not find the final upper device from
>> start_dev's upper list in bond_verify_device_path(), only the loopback
>> detection of the current upper device should be affected, and the system is
>> no need to be panic.
>> So return -ENOMEM in alb_upper_dev_walk to stop walking, print some warn
>> information when failed to allocate memory for vlan tags in
>> bond_verify_device_path.
>>
>> I also think that the following function calls
>> netdev_walk_all_upper_dev_rcu
>> ---->>>alb_upper_dev_walk
>> ---------->>>bond_verify_device_path
>>>  From this way, "end device" can eventually be obtained from "start device"
>> in bond_verify_device_path, IS_ERR(tags) could be instead of
>> IS_ERR_OR_NULL(tags) in alb_upper_dev_walk.
>>
>> Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
>> ---
>> v4: print information instead of warn
>> v3: return -ENOMEM instead of zero to stop walk
>> v2: use WARN_ON_ONCE instead of WARN_ON
>> ---
>>   drivers/net/bonding/bond_alb.c  | 3 ++-
>>   drivers/net/bonding/bond_main.c | 5 ++++-
>>   2 files changed, 6 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
>> index dc2c7b979656..7edf0fd58c34 100644
>> --- a/drivers/net/bonding/bond_alb.c
>> +++ b/drivers/net/bonding/bond_alb.c
>> @@ -985,7 +985,8 @@ static int alb_upper_dev_walk(struct net_device *upper,
>>   	if (netif_is_macvlan(upper) && !strict_match) {
>>   		tags = bond_verify_device_path(bond->dev, upper, 0);
>>   		if (IS_ERR_OR_NULL(tags))
>> -			BUG();
>> +			return -ENOMEM;
>> +
>>   		alb_send_lp_vid(slave, upper->dev_addr,
>>   				tags[0].vlan_proto, tags[0].vlan_id);
>>   		kfree(tags);
>> diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
>> index 51d47eda1c87..1a40bd08f984 100644
>> --- a/drivers/net/bonding/bond_main.c
>> +++ b/drivers/net/bonding/bond_main.c
>> @@ -2967,8 +2967,11 @@ struct bond_vlan_tag *bond_verify_device_path(struct net_device *start_dev,
>>   
>>   	if (start_dev == end_dev) {
>>   		tags = kcalloc(level + 1, sizeof(*tags), GFP_ATOMIC);
>> -		if (!tags)
>> +		if (!tags) {
>> +			net_err_ratelimited("%s: %s: Failed to allocate tags\n",
>> +					    __func__, start_dev->name);
> 
> I'm sorry for the repeated back and forth, but checkpatch points out
> that the above warning adds little value: if the memory allocation
> fails, the mm layer will emit a lot warning comprising the backtrace.
> 
> I suggest to drop it, thanks!
Thank you for your advice. And I will send v5 later.

Zhengchao Shao
> 
> Paolo
>