Re: [PATCH] io_uring: add napi busy settings to the fdinfo output

[Olivier Langlois <olivier@trillion01.com>]

Thank you Pavel for your review!

Since I have no indication if Jens did see your comment before applying
my patch, I will prepare another one with your comment addressed.

just ignore it if it is not needed anymore.

On Tue, 2024-07-30 at 12:05 +0100, Pavel Begunkov wrote:
> On 7/29/24 23:38, Olivier Langlois wrote:
> > this info may be useful when attempting to debug a problem
> > involving a ring using the feature.
> 
> Apart from a comment below,
> 
> Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
> 
> Maybe, Jens would be willing to move the block after the spin_unlock
> while applying.
> 
> 
> > Signed-off-by: Olivier Langlois <olivier@trillion01.com>
> > ---
> >   io_uring/fdinfo.c | 13 ++++++++++++-
> >   1 file changed, 12 insertions(+), 1 deletion(-)
> > 
> > diff --git a/io_uring/fdinfo.c b/io_uring/fdinfo.c
> > index b1e0e0d85349..3ba42e136a40 100644
> > --- a/io_uring/fdinfo.c
> > +++ b/io_uring/fdinfo.c
> > @@ -221,7 +221,18 @@ __cold void io_uring_show_fdinfo(struct
> > seq_file *m, struct file *file)
> >   			   cqe->user_data, cqe->res, cqe->flags);
> >   
> >   	}
> > -
> > +#ifdef CONFIG_NET_RX_BUSY_POLL
> > +	if (ctx->napi_enabled) {
> > +		seq_puts(m, "NAPI:\tenabled\n");
> > +		seq_printf(m, "napi_busy_poll_dt:\t%llu\n", ctx-
> > >napi_busy_poll_dt);
> > +		if (ctx->napi_prefer_busy_poll)
> > +			seq_puts(m,
> > "napi_prefer_busy_poll:\ttrue\n");
> > +		else
> > +			seq_puts(m,
> > "napi_prefer_busy_poll:\tfalse\n");
> > +	} else {
> > +		seq_puts(m, "NAPI:\tdisabled\n");
> > +	}
> > +#endif
> >   	spin_unlock(&ctx->completion_lock);
> 
> That doesn't need to be under completion_lock, it should move outside
> of the spin section.
> 
> 
> >   }
> >   #endif
>