[dpdk-dev] [PATCH 14/33] app/testeventdev: order: add eventdev port setup

Van Haaren, Harry harry.van.haaren at intel.com
Fri Jun 23 14:36:47 CEST 2017


> From: Jerin Jacob [mailto:jerin.jacob at caviumnetworks.com]
> Sent: Sunday, May 28, 2017 8:59 PM
> To: dev at dpdk.org
> Cc: Richardson, Bruce <bruce.richardson at intel.com>; Van Haaren, Harry
> <harry.van.haaren at intel.com>; hemant.agrawal at nxp.com; Eads, Gage <gage.eads at intel.com>;
> nipun.gupta at nxp.com; Vangati, Narender <narender.vangati at intel.com>; Rao, Nikhil
> <nikhil.rao at intel.com>; gprathyusha at caviumnetworks.com; Jerin Jacob
> <jerin.jacob at caviumnetworks.com>
> Subject: [dpdk-dev] [PATCH 14/33] app/testeventdev: order: add eventdev port setup
> 
> Setup one port per worker and link to all queues and setup
> one producer port to inject the events.
> 
> Signed-off-by: Jerin Jacob <jerin.jacob at caviumnetworks.com>

One suggestion below to increase the enqueue depth, or was there a reason set it to 8?

Acked-by: Harry van Haaren <harry.van.haaren at intel.com>


<snip some code>

> +	/* setup one port per worker, linking to all queues */
> +	for (port = 0; port < nb_workers; port++) {
> +		struct worker_data *w = &t->worker[port];
> +
> +		w->dev_id = opt->dev_id;
> +		w->port_id = port;
> +		w->t = t;
> +
> +		ret = rte_event_port_setup(opt->dev_id, port, &wkr_p_conf);
> +		if (ret) {
> +			evt_err("failed to setup port %d", port);
> +			return ret;
> +		}
> +
> +		ret = rte_event_port_link(opt->dev_id, port, NULL, NULL, 0);
> +		if (ret != nb_queues) {
> +			evt_err("failed to link all queues to port %d", port);
> +			return -EINVAL;
> +		}
> +	}
> +	/* port for producer, no links */
> +	const struct rte_event_port_conf prod_conf = {
> +			.dequeue_depth = 8,
> +			.enqueue_depth = 8,

.enqueue_depth = 32 or so?



More information about the dev mailing list