[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <9a0384e6-f3ec-1f43-0e1a-b0da30f3a7a5@mojatatu.com>
Date: Sun, 25 Sep 2016 07:32:42 -0400
From: Jamal Hadi Salim <jhs@...atatu.com>
To: Jesper Dangaard Brouer <brouer@...hat.com>
Cc: Tom Herbert <tom@...bertland.com>, davem@...emloft.net,
netdev@...r.kernel.org, kernel-team@...com, tariqt@...lanox.com,
bblanco@...mgrid.com, alexei.starovoitov@...il.com,
eric.dumazet@...il.com, Thomas Graf <tgraf@...g.ch>
Subject: Re: [PATCH RFC 1/3] xdp: Infrastructure to generalize XDP
On 16-09-23 09:00 AM, Jesper Dangaard Brouer wrote:
> On Fri, 23 Sep 2016 07:13:30 -0400
> Jamal Hadi Salim <jhs@...atatu.com> wrote:
>
[..]
>> Tom,
>> perused the thread and it seems you are serious ;->
>> Are we heading towards Frankenstein Avenue?
>> The whole point behind letting in XDP is so that _small programs_
>> can be written to do some quick thing. eBPF 4K program limit was
>> touted as the check and bound. eBPF sounded fine.
>> This sounds like a huge contradiction.
>>
>> cheers,
>> jamal
>
> Hi Jamal,
>
> I don't understand why you think this is so controversial. The way I
> see it (after reading the thread): This is about allowing kernel
> components to _also_ use the XDP hook.
>
The initial push was XDP to support very small programs that did
very simple things fast (to be extreme: for example running a whole
network stack is a no-no). EBPF with the 4K program limit was pointed
as the limit.
What Tom is presenting is implying this constraint is now being
removed. Thats the controversy.
cheers,
jamal
Powered by blists - more mailing lists