On Thu, Nov 29, 2018 at 3:27 AM Paolo Abeni wrote:
>
> Hi,
>
> Thank you for the update!
>
> On Wed, 2018-11-28 at 18:50 -0500, Willem de Bruijn wrote:
> > I did revert to the basic implementation using an extra ref
> > for the function call, similar to TCP, as you suggested.
> >
> > On top of tha
Hi,
Thank you for the update!
On Wed, 2018-11-28 at 18:50 -0500, Willem de Bruijn wrote:
> I did revert to the basic implementation using an extra ref
> for the function call, similar to TCP, as you suggested.
>
> On top of that as a separate optimization patch I have a
> variant that uses refcn
On Mon, Nov 26, 2018 at 2:49 PM Willem de Bruijn
wrote:
>
> On Mon, Nov 26, 2018 at 1:19 PM Willem de Bruijn
> wrote:
> >
> > On Mon, Nov 26, 2018 at 1:04 PM Paolo Abeni wrote:
> > >
> > > On Mon, 2018-11-26 at 12:59 -0500, Willem de Bruijn wrote:
> > > > The callers of this function do flush th
On Mon, Nov 26, 2018 at 1:19 PM Willem de Bruijn
wrote:
>
> On Mon, Nov 26, 2018 at 1:04 PM Paolo Abeni 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
On Mon, Nov 26, 2018 at 1:04 PM Paolo Abeni 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
On Mon, Nov 26, 2018 at 11:32 AM Paolo Abeni wrote:
>
> Hi,
>
> Sorry for the long delay...
>
> On Mon, 2018-11-26 at 10:29 -0500, Willem de Bruijn wrote:
> > @@ -1109,6 +1128,7 @@ static int __ip_append_data(struct sock *sk,
> > error_efault:
> > err = -EFAULT;
> > error:
> > + sock_z
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 c
Hi,
Sorry for the long delay...
On Mon, 2018-11-26 at 10:29 -0500, Willem de Bruijn wrote:
> @@ -1109,6 +1128,7 @@ static int __ip_append_data(struct sock *sk,
> error_efault:
> err = -EFAULT;
> error:
> + sock_zerocopy_put_abort(uarg);
> cork->length -= length;
> IP_INC_S
From: Willem de Bruijn
Extend zerocopy to udp sockets. Allow setting sockopt SO_ZEROCOPY and
interpret flag MSG_ZEROCOPY.
This patch was previously part of the zerocopy RFC patchsets. Zerocopy
is not effective at small MTU. With segmentation offload building
larger datagrams, the benefit of page
