[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAEivzxeZKUkL2-gSeK8R+PJ2NppB-OKcVYbvLbX-uKbtahf1SA@mail.gmail.com>
Date: Mon, 27 Feb 2023 11:21:13 +0100
From: Aleksandr Mikhalitsyn <aleksandr.mikhalitsyn@...onical.com>
To: Eric Dumazet <edumazet@...gle.com>
Cc: davem@...emloft.net, linux-kernel@...r.kernel.org,
netdev@...r.kernel.org, Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>
Subject: Re: [PATCH net-next] scm: fix MSG_CTRUNC setting condition for SO_PASSSEC
On Mon, Feb 27, 2023 at 11:01 AM Eric Dumazet <edumazet@...gle.com> wrote:
>
> On Sun, Feb 26, 2023 at 9:17 PM Alexander Mikhalitsyn
> <aleksandr.mikhalitsyn@...onical.com> wrote:
> >
> > Currently, we set MSG_CTRUNC flag is we have no
> > msg_control buffer provided and SO_PASSCRED is set
> > or if we have pending SCM_RIGHTS.
> >
> > For some reason we have no corresponding check for
> > SO_PASSSEC.
>
Hi Eric,
> Can you describe what side effects this patch has ?
>
> I think it could break some applications, who might not be able to
> recover from MSG_CTRUNC in this case.
> This should be documented, in order to avoid a future revert.
Yes, it can break applications but only those who use SO_PASSSEC
and not properly check MSG_CTRUNC. According to the recv(2) man:
MSG_CTRUNC
indicates that some control data was discarded due to lack
of space in the buffer for ancillary data.
So, there is no specification about a particular SCM type. It seems more correct
to handle SCM_SECURITY the same way as SCM_RIGHTS / SCM_CREDENTIALS.
>
> In any case, net-next is currently closed.
Oh, I'm sorry.
Kind regards,
Alex
Powered by blists - more mailing lists