> > > Subject: Re: [dpdk-dev] [PATCH v2 1/2] virtio: one way barrier for
> > > split vring used idx
> > >
> > > On Thu, 2 Apr 2020 10:57:52 +0800
> > > Joyce Kong wrote:
> > >
> > > > -(vq)->vq_used_cons_idx))
> >
> > Subject: Re: [dpdk-dev] [PATCH v2 1/2] virtio: one way barrier for
> > split vring used idx
> >
> > On Thu, 2 Apr 2020 10:57:52 +0800
> > Joyce Kong wrote:
> >
> > > -(vq)->vq_used_cons_idx))
> > > +static inline uint16_t
> > &g
Honnappa Nagarahalli
> ; Gavin Hu ; nd
> ; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 1/2] virtio: one way barrier for split vring
> used idx
>
> On Thu, 2 Apr 2020 10:57:52 +0800
> Joyce Kong wrote:
>
> > - (
On Thu, 2 Apr 2020 10:57:52 +0800
Joyce Kong wrote:
> - (vq)->vq_used_cons_idx))
> +static inline uint16_t
> +virtqueue_nused(struct virtqueue *vq)
vq is unmodified and should be const
> +{
> + uint16_t idx;
> + if (vq->hw->weak_barriers) {
Put blank
In case VIRTIO_F_ORDER_PLATFORM(36) is not negotiated, then the frontend
and backend are assumed to be implemented in software, that is they can
run on identical CPUs in an SMP configuration.
Thus a weak form of memory barriers like rte_smp_r/wmb, other than
rte_cio_r/wmb, is sufficient for this ca
5 matches
Mail list logo
