netdev - Re: [PATCH net-next 1/3] net: ipv6: ioam6_iptunnel: mitigate 2-realloc issue

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: <d3bce110-4b1b-44ed-8c1d-a9736a02f1dd@uliege.be>
Date: Fri, 25 Oct 2024 23:06:36 +0200
From: Justin Iurman <justin.iurman@...ege.be>
To: Alexander Lobakin <aleksander.lobakin@...el.com>
Cc: netdev@...r.kernel.org, davem@...emloft.net, dsahern@...nel.org,
 edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com, horms@...nel.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next 1/3] net: ipv6: ioam6_iptunnel: mitigate
 2-realloc issue

On 10/25/24 17:12, Alexander Lobakin wrote:
> From: Justin Iurman <justin.iurman@...ege.be>
> Date: Fri, 25 Oct 2024 15:37:25 +0200
> 
>> This patch mitigates the two-reallocations issue with ioam6_iptunnel by
>> providing the dst_entry (in the cache) to the first call to
>> skb_cow_head(). As a result, the very first iteration would still
>> trigger two reallocations (i.e., empty cache), while next iterations
>> would only trigger a single reallocation.
> 
> [...]
> 
>>   static int ioam6_do_inline(struct net *net, struct sk_buff *skb,
>> -			   struct ioam6_lwt_encap *tuninfo)
>> +			   struct ioam6_lwt_encap *tuninfo,
>> +			   struct dst_entry *dst)
>>   {
>>   	struct ipv6hdr *oldhdr, *hdr;
>>   	int hdrlen, err;
>>   
>>   	hdrlen = (tuninfo->eh.hdrlen + 1) << 3;
>>   
>> -	err = skb_cow_head(skb, hdrlen + skb->mac_len);
>> +	err = skb_cow_head(skb, hdrlen + (!dst ? skb->mac_len
>> +					       : LL_RESERVED_SPACE(dst->dev)));
> 
> You use this pattern a lot throughout the series. I believe you should
> make a static inline or a macro from it.
> 
> static inline u32 some_name(const *dst, const *skb)
> {
> 	return dst ? LL_RESERVED_SPACE(dst->dev) : skb->mac_len;
> }
> 
> BTW why do you check for `!dst`, not `dst`? Does changing this affects
> performance?

Not at all, you're right... even the opposite actually. Regarding the 
static inline suggestion, it could be a good idea and may even look like 
this as an optimization:

static inline u32 dev_overhead(struct dst_entry *dst, struct sk_buff *skb)
{
	if (likely(dst))
		return LL_RESERVED_SPACE(dst->dev);

	return skb->mac_len;
}

The question is... where should it go then? A static inline function per 
file (i.e., ioam6_iptunnel.c, seg6_iptunnel.c, and rpl_iptunnel.c)? In 
that case, it would still be repeated 3 times. Or in a header file 
somewhere, to have it defined only once? If so, what location do you 
think would be best?