> From: Pavan Nikhilesh [mailto:pbhagavat...@caviumnetworks.com]
> Sent: Wednesday, January 10, 2018 2:52 PM
> To: jerin.ja...@caviumnetworks.com; santosh.shu...@caviumnetworks.com; Van
> Haaren, Harry <harry.van.haa...@intel.com>; Eads, Gage
> <gage.e...@intel.com>; hemant.agra...@nxp.com; nipun.gu...@nxp.com; Ma,
> Liang J <liang.j...@intel.com>
> Cc: dev@dpdk.org; Pavan Nikhilesh <pbhagavat...@caviumnetworks.com>
> Subject: [dpdk-dev] [PATCH v3 09/12] app/eventdev: add pipeline queue worker
> functions
>
<snip>
> +static __rte_always_inline void
> +pipeline_tx_pkt_safe(struct rte_mbuf *mbuf)
> +{
> + while (rte_eth_tx_burst(mbuf->port, 0, &mbuf, 1) != 1)
> + rte_pause();
> +}
re safe, see comment below
> +
> +static __rte_always_inline void
> +pipeline_tx_pkt_unsafe(struct rte_mbuf *mbuf, struct test_pipeline *t)
> +{
> + rte_spinlock_t *lk = &t->tx_lk[mbuf->port];
> +
> + rte_spinlock_lock(lk);
> + pipeline_tx_pkt_safe(mbuf);
> + rte_spinlock_unlock(lk);
> +}
IIRC usually the "Safe" version of a function has extra locks/protection, while
the "normal" version has better performance, but less-error-checking.
Here, the "unsafe" function does the extra locking. If looking from the HW POV,
that makes sense, but I think its inverted from most existing code...
Happy to be proved wrong here .. ?
<snip>
> +static int
> +pipeline_queue_worker_single_stage_safe(void *arg)
> +{
> + struct worker_data *w = arg;
> + struct test_pipeline *t = w->t;
> + const uint8_t dev = w->dev_id;
> + const uint8_t port = w->port_id;
> + struct rte_event ev;
> +
> + while (t->done == false) {
> + uint16_t event = rte_event_dequeue_burst(dev, port, &ev, 1, 0);
> +
> + if (!event) {
> + rte_pause();
> + continue;
> + }
> +
> + if (ev.sched_type == RTE_SCHED_TYPE_ATOMIC) {
> + pipeline_tx_pkt_safe(ev.mbuf);
I guess that means that the functions where they're used are inverted in name
too.
<snip>
