netdev - Re: [PATCH netnext] mpls: Reduce skb re-allocations due to skb

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: <967C2745-1EDB-464E-9C80-46345CA91650@apple.com>
Date: Fri, 16 Aug 2024 15:20:03 -0700
From: Christoph Paasch <cpaasch@...le.com>
To: Simon Horman <horms@...nel.org>
Cc: netdev <netdev@...r.kernel.org>, Eric Dumazet <edumazet@...gle.com>,
 Paolo Abeni <pabeni@...hat.com>, Jakub Kicinski <kuba@...nel.org>,
 David Miller <davem@...emloft.net>, Roopa Prabhu <roopa@...dia.com>,
 Craig Taylor <cmtaylor@...le.com>
Subject: Re: [PATCH netnext] mpls: Reduce skb re-allocations due to skb_cow()

Hello!

> On Aug 16, 2024, at 4:18 AM, Simon Horman <horms@...nel.org> wrote:
> 
> On Thu, Aug 15, 2024 at 09:12:01AM -0700, Christoph Paasch wrote:
>> mpls_xmit() needs to prepend the MPLS-labels to the packet. That implies
>> one needs to make sure there is enough space for it in the headers.
>> 
>> Calling skb_cow() implies however that one wants to change even the
>> playload part of the packet (which is not true for MPLS). Thus, call
>> skb_cow_head() instead, which is what other tunnelling protocols do.
>> 
>> Running a server with this comm it entirely removed the calls to
>> pskb_expand_head() from the callstack in mpls_xmit() thus having
>> significant CPU-reduction, especially at peak times.
> 
> Hi Christoph and Craig,
> 
> Including some performance data here would be nice.

Getting exact production performance data is going to be a major challenge. Not a technical challenge, but rather logistically, ...

>> Cc: Roopa Prabhu <roopa@...dia.com>
>> Reported-by: Craig Taylor <cmtaylor@...le.com>
>> Signed-off-by: Christoph Paasch <cpaasch@...le.com>
> 
> 
> And one minor nit, which I do not think warrants a repost:
> netnext -> net-next.

Ugh - I was wondering why did patchwork thought it was a fix… that explains it.

If a respin is needed, please let me know.

> In any case, this looks good to me.
> 
> Reviewed-by: Simon Horman <horms@...nel.org>


Thanks,
Christoph