netdev - Re: [PATCH net] net/ipv6: Revert attempt to simplify route replace and append

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] [day] [month] [year] [list]

Message-ID: <7dcb15ac-9089-a323-1beb-2c07a61df242@gmail.com>
Date:   Wed, 4 Jul 2018 21:01:52 -0600
From:   David Ahern <dsahern@...il.com>
To:     David Miller <davem@...emloft.net>, idosch@...lanox.com
Cc:     dsahern@...nel.org, netdev@...r.kernel.org,
        sharpd@...ulusnetworks.com, Thomas.Winter@...iedtelesis.co.nz,
        petrm@...lanox.com
Subject: Re: [PATCH net] net/ipv6: Revert attempt to simplify route replace
 and append

On 7/4/18 8:29 PM, David Miller wrote:
> From: Ido Schimmel <idosch@...lanox.com>
> Date: Thu, 5 Jul 2018 00:10:41 +0300
> 
>> We can have the IPv4/IPv6 code only generate a REPLACE / DELETE
>> notification for routes that are actually used for forwarding and
>> relieve listeners from the need to implement this logic themselves. I
>> think this should work.
> 
> Whilst this could reduce the duplication, I worry that in the long
> term that this might end up being error prone.
> 

Duplication of data and duplication of logic is not ideal. Especially in
this case where the duplication of both is only to handle one case -
duplicate routes where only the first is programmed. I suspect it will
have to be dealt with at some point (e.g., scaling to a million routes),
but right now there are more important factors to deal with - like the
rtnl_lock. Something to keep in mind for the future.