RE: [PATCH 2/2] net/nfp: enhance the flower service framework

[Chaoyong He <chaoyong.he@corigine.com>]

> On Thu, 10 Oct 2024 15:45:57 +0800
> Chaoyong He <chaoyong.he@corigine.com> wrote:
> 
> > +static void
> > +nfp_flower_service_alarm_func(void *arg) {
> > +	int ret;
> > +	uint16_t slot;
> > +	struct nfp_net_hw_priv *hw_priv;
> > +	struct nfp_flower_service *service_handle;
> > +
> > +	service_handle = arg;
> > +	if (!service_handle->alarm_enabled)
> > +		goto alarm_set;
> > +
> > +	rte_spinlock_lock(&service_handle->spinlock);
> > +	for (slot = 0; slot < MAX_FLOWER_SERVICE_SLOT; slot++) {
> > +		hw_priv = service_handle->slots[slot];
> > +		if (hw_priv == NULL)
> > +			continue;
> > +
> > +		nfp_flower_ctrl_vnic_process(hw_priv);
> > +	}
> > +	rte_spinlock_unlock(&service_handle->spinlock);
> > +
> 
> The alarm handling is in a non EAL thread, and service is an lcore; does that
> matter to this code?

It does not matter to this code, maybe the name of this function and some variables introduced a few doubts, but the logic is okay.
Thanks for the review.