Re: [PATCH net-next 5/5] net: optimize napi_threaded_poll() vs RPS/RFS

[Paolo Abeni <pabeni@redhat.com>]

On Fri, 2023-04-21 at 16:06 +0200, Eric Dumazet wrote:
> On Fri, Apr 21, 2023 at 3:10 PM Paolo Abeni <pabeni@redhat.com> wrote:
> > 
> > Hello,
> > 
> > thank you for the extremely fast turnaround!
> > 
> > On Fri, 2023-04-21 at 09:43 +0000, Eric Dumazet wrote:
> > > We use napi_threaded_poll() in order to reduce our softirq dependency.
> > > 
> > > We can add a followup of 821eba962d95 ("net: optimize napi_schedule_rps()")
> > > to further remove the need of firing NET_RX_SOFTIRQ whenever
> > > RPS/RFS are used.
> > > 
> > > Signed-off-by: Eric Dumazet <edumazet@google.com>
> > > ---
> > >  include/linux/netdevice.h |  3 +++
> > >  net/core/dev.c            | 12 ++++++++++--
> > >  2 files changed, 13 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
> > > index a6a3e9457d6cbc9fcbbde96b43b4b21878495403..08fbd4622ccf731daaee34ad99773d6dc2e82fa6 100644
> > > --- a/include/linux/netdevice.h
> > > +++ b/include/linux/netdevice.h
> > > @@ -3194,7 +3194,10 @@ struct softnet_data {
> > >  #ifdef CONFIG_RPS
> > >       struct softnet_data     *rps_ipi_list;
> > >  #endif
> > > +
> > >       bool                    in_net_rx_action;
> > > +     bool                    in_napi_threaded_poll;
> > 
> > If I'm correct only one of the above 2 flags can be set to true at any
> > give time. I'm wondering if could use a single flag (possibly with a
> > rename - say 'in_napi_polling')?
> 
> Well, we can _not_ use the same flag, because we do not want to
> accidentally enable
> the part in ____napi_schedule()

I see, thanks for the pointer.
> 
> We could use a bit mask with 2 bits, but I am not sure it will help readability.

Agreed, it's better to use 2 separate bool.

LGTM,

Acked-by: Paolo Abeni <pabeni@redhat.com>

(for pw's sake ;)