[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <338121b9-5e71-be85-7faf-02dbba070df3@gmail.com>
Date: Fri, 24 Nov 2017 10:14:51 -0700
From: David Ahern <dsahern@...il.com>
To: Solio Sarabia <solio.sarabia@...el.com>, netdev@...r.kernel.org,
davem@...eloft.net, stephen@...workplumber.org
Cc: kys@...rosoft.com, shiny.sebastian@...el.com,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] net-sysfs: export gso_max_size attribute
On 11/22/17 5:30 PM, Solio Sarabia wrote:
> The netdevice gso_max_size is exposed to allow users fine-control on
> systems with multiple NICs with different GSO buffer sizes, and where
> the virtual devices like bridge and veth, need to be aware of the GSO
> size of the underlying devices.
>
> In a virtualized environment, setting the right GSO sizes for physical
> and virtual devices makes all TSO work to be on physical NIC, improving
> throughput and reducing CPU util. If virtual devices send buffers
> greater than what NIC supports, it forces host to do TSO for buffers
> exceeding the limit, increasing CPU utilization in host.
>
> Suggested-by: Shiny Sebastian <shiny.sebastian@...el.com>
> Signed-off-by: Solio Sarabia <solio.sarabia@...el.com>
> ---
This should be added to rtnetlink rather than sysfs.
Powered by blists - more mailing lists