[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5e968e6d-4dbf-cfef-a270-99b0d1dd4ea8@redhat.com>
Date: Wed, 2 May 2018 13:17:50 -0400
From: Vlad Yasevich <vyasevic@...hat.com>
To: Marcelo Ricardo Leitner <marcelo.leitner@...il.com>,
Vladislav Yasevich <vyasevich@...il.com>
Cc: netdev@...r.kernel.org, linux-sctp@...r.kernel.org,
virtualization@...ts.linux-foundation.org,
virtio-dev@...ts.oasis-open.org, mst@...hat.com,
jasowang@...hat.com, nhorman@...driver.com
Subject: Re: [PATCH V2 net-next 4/6] tun: Add support for SCTP checksum
offload
On 05/02/2018 10:56 AM, Marcelo Ricardo Leitner wrote:
> On Wed, May 02, 2018 at 11:53:47AM -0300, Marcelo Ricardo Leitner wrote:
>> On Tue, May 01, 2018 at 10:07:37PM -0400, Vladislav Yasevich wrote:
>>> Adds a new tun offload flag to allow for SCTP checksum offload.
>>> The flag has to be set by the user and defaults to "no offload".
>>
>> I'm confused here:
>>
>>> +++ b/drivers/net/tun.c
>>> @@ -216,7 +216,7 @@ struct tun_struct {
>>> struct net_device *dev;
>>> netdev_features_t set_features;
>>> #define TUN_USER_FEATURES (NETIF_F_HW_CSUM|NETIF_F_TSO_ECN|NETIF_F_TSO| \
>>> - NETIF_F_TSO6)
>>> + NETIF_F_TSO6|NETIF_F_SCTP_CRC)
>>
>> Doesn't adding it here mean it defaults to "offload", instead?
>>
>> later on, it does:
>> dev->hw_features = NETIF_F_SG | NETIF_F_FRAGLIST |
>> TUN_USER_FEATURES | NETIF_F_HW_VLAN_CTAG_TX |
>> NETIF_F_HW_VLAN_STAG_TX;
>
> Missed to paste the next line too:
> dev->features = dev->hw_features | NETIF_F_LLTX;
>
Yes, as a software device, we can default it to on. However, qemu will 0-out
the features and then set them up based on the guest (just like regular checksum).
-vlad
Powered by blists - more mailing lists