Re: [dpdk-dev] [PATCH v5 1/2] virtio: one way barrier for split vring used idx

2020-05-04 Thread Maxime Coquelin
On 5/1/20 12:58 AM, Ferruh Yigit wrote: > On 4/30/2020 10:14 AM, Joyce Kong wrote: >> 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 we

Re: [dpdk-dev] [PATCH v5 1/2] virtio: one way barrier for split vring used idx

2020-04-30 Thread Ferruh Yigit
On 4/30/2020 10:14 AM, Joyce Kong wrote: > 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, ot