[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200227091637.1822b9ab@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
Date: Thu, 27 Feb 2020 09:16:37 -0800
From: Jakub Kicinski <kuba@...nel.org>
To: "Michael S. Tsirkin" <mst@...hat.com>
Cc: David Ahern <dahern@...italocean.com>,
Toke Høiland-Jørgensen
<toke@...hat.com>, Jason Wang <jasowang@...hat.com>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
Michael Chan <michael.chan@...adcom.com>,
Andy Gospodarek <andy@...yhouse.net>
Subject: Re: virtio_net: can change MTU after installing program
On Thu, 27 Feb 2020 03:14:20 -0500 Michael S. Tsirkin wrote:
> On Wed, Feb 26, 2020 at 05:37:51PM -0800, Jakub Kicinski wrote:
> > On Wed, 26 Feb 2020 12:02:03 -0500 Michael S. Tsirkin wrote:
> > I'd vote that we don't care. We should care more about consistency
> > across drivers than committing to buggy behavior.
> >
> > All drivers should have this check (intel, mlx, nfp definitely do),
> > I had a look at Broadcom and it seems to be missing there as well :(
> > Qlogic also. Ugh.
>
> Any chance to put it in net core then? Seems straight-forward enough ...
It's not impossible, but generally the RX buf geometry requirements
are not universal (see ixgbe or nfp).
Powered by blists - more mailing lists