netdev - Re: Latest net-next from GIT panic

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 21 Sep 2017 15:18:53 +0200
From:   Paweł Staszewski <pstaszewski@...are.pl>
To:     Eric Dumazet <eric.dumazet@...il.com>
Cc:     Wei Wang <weiwan@...gle.com>, Cong Wang <xiyou.wangcong@...il.com>,
        Linux Kernel Network Developers <netdev@...r.kernel.org>,
        Eric Dumazet <edumazet@...gle.com>
Subject: Re: Latest net-next from GIT panic



W dniu 2017-09-21 o 13:31, Paweł Staszewski pisze:
>
>
> W dniu 2017-09-21 o 13:03, Eric Dumazet pisze:
>> OK we have two problems here
>>
>> 1) We need to unify skb_dst_force()  ( for net tree )
>>
>> 2) Vlan devices should try to correctly handle IFF_XMIT_DST_RELEASE from
>> lower device. This will considerably help your performance.
>>
>>
>> For 1), this is what I had in mind, can you try it ?
>>
>> Thanks a lot !
>>
>> diff --git a/include/net/dst.h b/include/net/dst.h
>> index 
>> 93568bd0a3520bb7402f04d90cf04ac99c81cfbe..f23851eeaad917e8dafc06b58d23a2575405c894 
>> 100644
>> --- a/include/net/dst.h
>> +++ b/include/net/dst.h
>> @@ -271,7 +271,7 @@ static inline void dst_use_noref(struct dst_entry 
>> *dst, unsigned long time)
>>   static inline struct dst_entry *dst_clone(struct dst_entry *dst)
>>   {
>>       if (dst)
>> -        atomic_inc(&dst->__refcnt);
>> +        dst_hold(dst);
>>       return dst;
>>   }
>>   @@ -311,21 +311,6 @@ static inline void skb_dst_copy(struct sk_buff 
>> *nskb, const struct sk_buff *oskb
>>       __skb_dst_copy(nskb, oskb->_skb_refdst);
>>   }
>>   -/**
>> - * skb_dst_force - makes sure skb dst is refcounted
>> - * @skb: buffer
>> - *
>> - * If dst is not yet refcounted, let's do it
>> - */
>> -static inline void skb_dst_force(struct sk_buff *skb)
>> -{
>> -    if (skb_dst_is_noref(skb)) {
>> -        WARN_ON(!rcu_read_lock_held());
>> -        skb->_skb_refdst &= ~SKB_DST_NOREF;
>> -        dst_clone(skb_dst(skb));
>> -    }
>> -}
>> -
>>   /**
>>    * dst_hold_safe - Take a reference on a dst if possible
>>    * @dst: pointer to dst entry
>> @@ -356,6 +341,23 @@ static inline void skb_dst_force_safe(struct 
>> sk_buff *skb)
>>       }
>>   }
>>   +/**
>> + * skb_dst_force - makes sure skb dst is refcounted
>> + * @skb: buffer
>> + *
>> + * If dst is not yet refcounted, let's do it
>> + */
>> +static inline void skb_dst_force(struct sk_buff *skb)
>> +{
>> +    if (skb_dst_is_noref(skb)) {
>> +        struct dst_entry *dst = skb_dst(skb);
>> +
>> +        WARN_ON(!rcu_read_lock_held());
>> +        if (!dst_hold_safe(dst))
>> +            dst = NULL;
>> +        skb->_skb_refdst = (unsigned long)dst;
>> +    }
>> +}
>>     /**
>>    *    __skb_tunnel_rx - prepare skb for rx reinsert
>>
>>
>
> Patch applied - soo far no problems - and no warnings in dmesg
>
>
ok after adding patch all is working from now for about 1 hour of normal 
traffic witc all bgp sessions connected and about 600k prefixes in kernel.