[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CANn89iJjP2c73jNveuCNuPsbp0tbH_zp3ciwF7D24_BbP8qV0g@mail.gmail.com>
Date: Tue, 7 May 2024 11:35:52 +0200
From: Eric Dumazet <edumazet@...gle.com>
To: Paolo Abeni <pabeni@...hat.com>
Cc: Simon Horman <horms@...nel.org>, "David S. Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>, netdev@...r.kernel.org, eric.dumazet@...il.com
Subject: Re: [PATCH net-next 3/8] rtnetlink: do not depend on RTNL for
IFLA_TXQLEN output
On Tue, May 7, 2024 at 11:26 AM Paolo Abeni <pabeni@...hat.com> wrote:
>
> On Sun, 2024-05-05 at 15:43 +0100, Simon Horman wrote:
> > On Fri, May 03, 2024 at 07:20:54PM +0000, Eric Dumazet wrote:
> > > rtnl_fill_ifinfo() can read dev->tx_queue_len locklessly,
> > > granted we add corresponding READ_ONCE()/WRITE_ONCE() annotations.
> > >
> > > Add missing READ_ONCE(dev->tx_queue_len) in teql_enqueue()
> >
> > Hi Eric,
> >
> > I am wondering if READ_ONCE(caifd->netdev->tx_queue_len)
> > is also missing from net/caif/caif_dev.c:transmit().
>
> I agree such read is outside the rtnl lock and could use a READ_ONCE
> annotation. I think it's better to handle that as an eventual follow-up
> instead of blocking this series.
I missed Simon feedback, sorry.
Yes, we can add missing READ_ONCE() as follow ups.
They are all orthogonal.
Powered by blists - more mailing lists