[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20210120104621.GM19605@breakpoint.cc>
Date: Wed, 20 Jan 2021 11:46:21 +0100
From: Florian Westphal <fw@...len.de>
To: menglong8.dong@...il.com
Cc: kuba@...nel.org, christian.brauner@...ntu.com, davem@...emloft.net,
yoshfuji@...ux-ipv6.org, dong.menglong@....com.cn,
daniel@...earbox.net, gnault@...hat.com, ast@...nel.org,
nicolas.dichtel@...nd.com, ap420073@...il.com, edumazet@...gle.com,
pabeni@...hat.com, jakub@...udflare.com, bjorn.topel@...el.com,
keescook@...omium.org, viro@...iv.linux.org.uk, rdna@...com,
maheshb@...gle.com, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next 3/3] net: core: Namespace-ify sysctl_rmem_max
and sysctl_wmem_max
menglong8.dong@...il.com <menglong8.dong@...il.com> wrote:
> From: Menglong Dong <dong.menglong@....com.cn>
>
> For now, sysctl_wmem_max and sysctl_rmem_max are globally unified.
> It's not convenient in some case. For example, when we use docker
> and try to control the default udp socket receive buffer for each
> container.
>
> For that reason, make sysctl_wmem_max and sysctl_rmem_max
> per-namespace.
I think having those values be restricted by init netns is a desirable
property.
Powered by blists - more mailing lists