[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <b4099188-cd5d-cbca-001b-3b0e4b2bb98a@mellanox.com>
Date: Mon, 13 Jul 2020 23:17:50 -0400
From: Ariel Levkovich <lariel@...lanox.com>
To: Cong Wang <xiyou.wangcong@...il.com>
Cc: Linux Kernel Network Developers <netdev@...r.kernel.org>,
Jiri Pirko <jiri@...nulli.us>,
Jakub Kicinski <kuba@...nel.org>,
Jamal Hadi Salim <jhs@...atatu.com>,
Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
Jiri Pirko <jiri@...lanox.com>
Subject: Re: [PATCH net-next v3 2/4] net/sched: Introduce action hash
On 7/13/20 6:04 PM, Cong Wang wrote:
> On Sat, Jul 11, 2020 at 2:28 PM Ariel Levkovich <lariel@...lanox.com> wrote:
>> Allow user to set a packet's hash value using a bpf program.
>>
>> The user provided BPF program is required to compute and return
>> a hash value for the packet which is then stored in skb->hash.
> Can be done by act_bpf, right?
Right. We already agreed on that.
Nevertheless, as I mentioned, act_bpf is not offloadable.
Device driver has no clue what the program does.
>
>> Using this action to set the skb->hash is an alternative to setting
>> it with act_skbedit and can be useful for future HW offload support
>> when the HW hash function is different then the kernel's hash
>> function.
>> By using a bpg program that emulates the HW hash function user
>> can ensure hash consistency between the SW and the HW.
> It sounds weird that the sole reason to add a new action is
> because of HW offloading. What prevents us extending the
> existing actions to support HW offloading?
>
> Thanks.
Something like adding a parameter to act_bpf that provides information
on the program
and what it does?
Ariel
Powered by blists - more mailing lists