[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1718762578.3916998-2-hengqi@linux.alibaba.com>
Date: Wed, 19 Jun 2024 10:02:58 +0800
From: Heng Qi <hengqi@...ux.alibaba.com>
To: Jakub Kicinski <kuba@...nel.org>
Cc: Jason Wang <jasowang@...hat.com>,
netdev@...r.kernel.org,
virtualization@...ts.linux.dev,
Thomas Huth <thuth@...ux.vnet.ibm.com>,
"Michael S. Tsirkin" <mst@...hat.com>,
Xuan Zhuo <xuanzhuo@...ux.alibaba.com>,
Eugenio PĂ©rez <eperezma@...hat.com>,
Eric Dumazet <edumazet@...gle.com>,
"David S. Miller" <davem@...emloft.net>,
Paolo Abeni <pabeni@...hat.com>,
Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
Jesper Dangaard Brouer <hawk@...nel.org>,
John Fastabend <john.fastabend@...il.com>
Subject: Re: [PATCH 1/2] virtio_net: checksum offloading handling fix
On Tue, 18 Jun 2024 18:15:16 -0700, Jakub Kicinski <kuba@...nel.org> wrote:
> On Tue, 18 Jun 2024 11:09:02 +0800 Heng Qi wrote:
> > > (Should we manually do checksum if RXCUSM is disabled?)
> > >
> >
> > Currently we do not allow RXCUSM to be disabled.
>
> You don't have to disable checksuming in the device.
Yes, it is up to the device itself to decide whether to validate checksum.
What I mean is that we don't allow users to disable the driver's
NETIF_F_RXCSUM flag.
> Just ignore VIRTIO_NET_HDR_F_DATA_VALID if user cleared NETIF_F_RXCSUM.
Right.
Thanks.
> I know some paranoid workloads which do actually want the kernel to
> calculate the checksum.
>
Powered by blists - more mailing lists