[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5e2652e4239a8_20912afc5c86e5c49f@john-XPS-13-9370.notmuch>
Date: Mon, 20 Jan 2020 17:24:52 -0800
From: John Fastabend <john.fastabend@...il.com>
To: Björn Töpel <bjorn.topel@...il.com>,
netdev@...r.kernel.org, ast@...nel.org, daniel@...earbox.net
Cc: Björn Töpel <bjorn.topel@...el.com>,
bpf@...r.kernel.org, magnus.karlsson@...el.com,
jonathan.lemon@...il.com
Subject: RE: [PATCH bpf-next] xsk: update rings for load-acquire/store-release
semantics
Björn Töpel wrote:
> From: Björn Töpel <bjorn.topel@...el.com>
>
> Currently, the AF_XDP rings uses fences for the kernel-side
> produce/consume functions. By updating rings for
> load-acquire/store-release semantics, the full barrier (smp_mb()) on
> the consumer side can be replaced.
>
> Signed-off-by: Björn Töpel <bjorn.topel@...el.com>
> ---
> net/xdp/xsk_queue.h | 31 ++++++++++++++-----------------
> 1 file changed, 14 insertions(+), 17 deletions(-)
>
> diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
> index bec2af11853a..2fff80576ee1 100644
> --- a/net/xdp/xsk_queue.h
> +++ b/net/xdp/xsk_queue.h
> @@ -39,19 +39,18 @@ struct xsk_queue {
> u64 invalid_descs;
> };
Should the xsk_cons_* libbpf routines also be updated then as well?
In general my understanding is the userspace and kernel should be
in sync?
Powered by blists - more mailing lists