From mboxrd@z Thu Jan  1 00:00:00 1970
From: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
Subject: Re: [PATCH net-next v2 1/2] udp: msg_zerocopy
Date: Mon, 26 Nov 2018 13:19:23 -0500
Message-ID: <CAF=yD-KG+wyHkX40FXPXNiNEHeKjOd5hPSDFuh-z0pUAnLehLg@mail.gmail.com>
References: <20181126152939.258443-1-willemdebruijn.kernel@gmail.com>
 <20181126152939.258443-2-willemdebruijn.kernel@gmail.com> <11350ff03fc6b03e34f8b4acd063371c887758d8.camel@redhat.com>
 <CAF=yD-J-QP_zUFY9W7CMdjOPgAwkHGgSfVQfVb=EKTNOZB4P8Q@mail.gmail.com> <ae1b8e52272b14a49ff4225e647c86c0aab57980.camel@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Cc: Network Development <netdev@vger.kernel.org>,
        David Miller <davem@davemloft.net>,
        Willem de Bruijn <willemb@google.com>
To: Paolo Abeni <pabeni@redhat.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-ed1-f67.google.com ([209.85.208.67]:38378 "EHLO
        mail-ed1-f67.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1725884AbeK0FO4 (ORCPT
        <rfc822;netdev@vger.kernel.org>); Tue, 27 Nov 2018 00:14:56 -0500
Received: by mail-ed1-f67.google.com with SMTP id h50so16715185ede.5
        for <netdev@vger.kernel.org>; Mon, 26 Nov 2018 10:20:01 -0800 (PST)
In-Reply-To: <ae1b8e52272b14a49ff4225e647c86c0aab57980.camel@redhat.com>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Mon, Nov 26, 2018 at 1:04 PM Paolo Abeni <pabeni@redhat.com> wrote:
>
> On Mon, 2018-11-26 at 12:59 -0500, Willem de Bruijn wrote:
> > The callers of this function do flush the queue of the other skbs on
> > error, but only after the call to sock_zerocopy_put_abort.
> >
> > sock_zerocopy_put_abort depends on total rollback to revert the
> > sk_zckey increment and suppress the completion notification (which
> > must not happen on return with error).
> >
> > I don't immediately have a fix. Need to think about this some more..
>
> [still out of sheer ignorance] How about tacking a refcnt for the whole
> ip_append_data() scope, like in the tcp case? that will add an atomic
> op per loop (likely, hitting the cache) but will remove some code hunk
> in sock_zerocopy_put_abort() and sock_zerocopy_alloc().

The atomic op pair is indeed what I was trying to avoid. But I also need
to solve the problem that the final decrement will happen from the freeing
of the other skbs in __ip_flush_pending_frames, and will not suppress
the notification.

Freeing the entire queue inside __ip_append_data, effectively making it
a true noop on error is one approach. But that is invasive, also to non
zerocopy codepaths, so I would rather avoid that.

Perhaps I need to handle the abort logic in udp_sendmsg directly,
after both __ip_append_data and __ip_flush_pending_frames.