netdev - Re: [PATCH net-next] ipv6: sit: Add ipip6_tunnel_dst

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]

Message-ID: <83aeba1c-27ff-4a92-b91e-5719be879e7a@huawei.com>
Date: Thu, 28 Aug 2025 14:44:07 +0800
From: Yue Haibing <yuehaibing@...wei.com>
To: Alexander Lobakin <aleksander.lobakin@...el.com>
CC: <davem@...emloft.net>, <dsahern@...nel.org>, <edumazet@...gle.com>,
	<kuba@...nel.org>, <pabeni@...hat.com>, <horms@...nel.org>,
	<netdev@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH net-next] ipv6: sit: Add ipip6_tunnel_dst_find() for
 cleanup

On 2025/8/28 14:13, Yue Haibing wrote:
> On 2025/8/27 22:45, Alexander Lobakin wrote:
>> From: Yue Haibing <yuehaibing@...wei.com>
>> Date: Wed, 27 Aug 2025 12:00:27 +0800
>>
>>> Extract the dst lookup logic from ipip6_tunnel_xmit() into new helper
>>> ipip6_tunnel_dst_find() to reduce code duplication and enhance readability.
>>>
>>> No functional change intended.
>>
>> ...but bloat-o-meter stats would be nice to see here. I'm curious
>> whether the object code got any changes or the compilers still just
>> inline this function to both call sites.

On a x86_64, with allmodconfig:

./scripts/bloat-o-meter -c net/ipv6/sit.o net/ipv6/sit-after.o
add/remove: 2/0 grow/shrink: 1/1 up/down: 1770/-2152 (-382)
Function                                     old     new   delta
ipip6_tunnel_dst_find                          -    1697   +1697
__pfx_ipip6_tunnel_dst_find                    -      64     +64
ipip6_tunnel_xmit.isra.cold                   79      88      +9
ipip6_tunnel_xmit.isra                      9910    7758   -2152
Total: Before=70060, After=69678, chg -0.55%
add/remove: 2/2 grow/shrink: 0/1 up/down: 16/-72 (-56)
Data                                         old     new   delta
__UNIQUE_ID___addressable_init_module2092       -       8      +8
__UNIQUE_ID___addressable_cleanup_module2093       -       8      +8
__UNIQUE_ID___addressable_init_module2093       8       -      -8
__UNIQUE_ID___addressable_cleanup_module2094       8       -      -8
descriptor                                   112      56     -56
Total: Before=752, After=696, chg -7.45%
add/remove: 1/1 grow/shrink: 2/2 up/down: 55/-51 (4)
RO Data                                      old     new   delta
__UNIQUE_ID_modinfo2094                        -      43     +43
__UNIQUE_ID_modinfo2096                       12      20      +8
__func__                                      55      59      +4
__UNIQUE_ID_modinfo2097                       20      18      -2
__UNIQUE_ID_modinfo2098                       18       -     -18
__UNIQUE_ID_modinfo2095                       43      12     -31
Total: Before=1725, After=1729, chg +0.23%


> 
> On a x86_64, with allmodconfig:
> before:
>    text    data     bss     dec     hex filename
>   96973   14687     256  111916   1b52c net/ipv6/sit.o
> 
> after:
>    text    data     bss     dec     hex filename
>   96699   14599     256  111554   1b3c2 net/ipv6/sit.o
> 
> I will add this in v2.
> 
>>
>>>
>>> Signed-off-by: Yue Haibing <yuehaibing@...wei.com>
>>> ---
>>>  net/ipv6/sit.c | 93 +++++++++++++++++++++++---------------------------
>>>  1 file changed, 43 insertions(+), 50 deletions(-)
>>>
>>> diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
>>> index 12496ba1b7d4..bcd261ff985b 100644
>>> --- a/net/ipv6/sit.c
>>> +++ b/net/ipv6/sit.c
>>> @@ -848,6 +848,47 @@ static inline __be32 try_6rd(struct ip_tunnel *tunnel,
>>>  	return dst;
>>>  }
>>>  
>>> +static bool ipip6_tunnel_dst_find(struct sk_buff *skb, __be32 *dst,
>>> +				  bool is_isatap)
>>> +{
>>> +	const struct ipv6hdr *iph6 = ipv6_hdr(skb);
>>> +	struct neighbour *neigh = NULL;
>>> +	const struct in6_addr *addr6;
>>> +	bool found = false;
>>> +	int addr_type;
>>> +
>>> +	if (skb_dst(skb))
>>> +		neigh = dst_neigh_lookup(skb_dst(skb), &iph6->daddr);
>>> +
>>> +	if (!neigh) {
>>> +		net_dbg_ratelimited("nexthop == NULL\n");
>>> +		return found;
>>
>> Return false here right away.
>>
>>> +	}
>>> +
>>> +	addr6 = (const struct in6_addr *)&neigh->primary_key;
>>> +	addr_type = ipv6_addr_type(addr6);
>>> +
>>> +	if (is_isatap) {
>>> +		if ((addr_type & IPV6_ADDR_UNICAST) &&
>>> +		    ipv6_addr_is_isatap(addr6)) {
>>> +			*dst = addr6->s6_addr32[3];
>>> +			found = true;
>>> +		}
>>> +	} else {
>>> +		if (addr_type == IPV6_ADDR_ANY) {
>>> +			addr6 = &ipv6_hdr(skb)->daddr;
>>> +			addr_type = ipv6_addr_type(addr6);
>>> +		}
>>> +
>>> +		if ((addr_type & IPV6_ADDR_COMPATv4) != 0) {
>>> +			*dst = addr6->s6_addr32[3];
>>> +			found = true;
>>> +		}
>>> +	}
>>> +	neigh_release(neigh);
>>> +	return found;
>>
>> I'd put 2 additional newlines here:
>>
>> 	}
>>
>> 	neigh_release(neigh);
>>
>> 	return found;
>> }
>>
>> for readability purposes and also a NL before the final `return` is
>> usually mandatory.
> 
> OK， thanks.
>>
>> Thanks,
>> Olek
>>
> 
>