[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <87o7dg3w48.fsf@toke.dk>
Date: Sat, 20 Jan 2024 15:41:43 +0100
From: Toke Høiland-Jørgensen <toke@...hat.com>
To: Jakub Kicinski <kuba@...nel.org>
Cc: Sebastian Andrzej Siewior <bigeasy@...utronix.de>, Alexei Starovoitov
<alexei.starovoitov@...il.com>, LKML <linux-kernel@...r.kernel.org>,
Network Development <netdev@...r.kernel.org>, "David S. Miller"
<davem@...emloft.net>, Boqun Feng <boqun.feng@...il.com>, Daniel Borkmann
<daniel@...earbox.net>, Eric Dumazet <edumazet@...gle.com>, Frederic
Weisbecker <frederic@...nel.org>, Ingo Molnar <mingo@...hat.com>, Paolo
Abeni <pabeni@...hat.com>, Peter Zijlstra <peterz@...radead.org>, Thomas
Gleixner <tglx@...utronix.de>, Waiman Long <longman@...hat.com>, Will
Deacon <will@...nel.org>, Alexei Starovoitov <ast@...nel.org>, Andrii
Nakryiko <andrii@...nel.org>, Cong Wang <xiyou.wangcong@...il.com>, Hao
Luo <haoluo@...gle.com>, Jamal Hadi Salim <jhs@...atatu.com>, Jesper
Dangaard Brouer <hawk@...nel.org>, Jiri Olsa <jolsa@...nel.org>, Jiri
Pirko <jiri@...nulli.us>, John Fastabend <john.fastabend@...il.com>, KP
Singh <kpsingh@...nel.org>, Martin KaFai Lau <martin.lau@...ux.dev>, Ronak
Doshi <doshir@...are.com>, Song Liu <song@...nel.org>, Stanislav Fomichev
<sdf@...gle.com>, VMware PV-Drivers Reviewers <pv-drivers@...are.com>,
Yonghong Song <yonghong.song@...ux.dev>, bpf <bpf@...r.kernel.org>
Subject: Re: [PATCH net-next 15/24] net: Use nested-BH locking for XDP
redirect.
Jakub Kicinski <kuba@...nel.org> writes:
> On Thu, 18 Jan 2024 12:51:18 +0100 Toke Høiland-Jørgensen wrote:
>> I do agree that conceptually it makes a lot of sense to encapsulate the
>> budget like this so drivers don't have to do all this state tracking
>> themselves. It does appear that drivers are doing different things with
>> the budget as it is today, though. For instance, the intel drivers seem
>> to divide the budget over all the enabled RX rings(?); so I'm wondering
>> if it'll be possible to unify drivers around a more opaque NAPI poll API?
>
> We can come up with APIs which would cater to multi-queue cases.
> Bigger question is what is the sensible polling strategy for those,
> just dividing the budget seems, hm, crude.
Right, agreed, though I don't have a good answer for what else to do off
the top of my head...
-Toke
Powered by blists - more mailing lists