[dpdk-dev] [RFC 1/7] net/af_xdp: new PMD driver

Zhang, Qi Z qi.z.zhang at intel.com
Thu Mar 1 02:51:08 CET 2018



> -----Original Message-----
> From: Stephen Hemminger [mailto:stephen at networkplumber.org]
> Sent: Thursday, March 1, 2018 7:42 AM
> To: Zhang, Qi Z <qi.z.zhang at intel.com>
> Cc: dev at dpdk.org; magnus.karlsson at intei.com; Topel, Bjorn
> <bjorn.topel at intel.com>
> Subject: Re: [dpdk-dev] [RFC 1/7] net/af_xdp: new PMD driver
> 
> On Tue, 27 Feb 2018 17:33:00 +0800
> Qi Zhang <qi.z.zhang at intel.com> wrote:
> 
> > struct pmd_internals {
> > +	int sfd;
> > +	int if_index;
> > +	char if_name[0x100];
> 
> why not IFNAMSIZ?
> 
> > +	struct ether_addr eth_addr;
> > +	struct xdp_queue rx;
> > +	struct xdp_queue tx;
> > +	struct xdp_umem *umem;
> > +	struct rte_mempool *mb_pool;
> > +
> > +	unsigned long rx_pkts;
> > +	unsigned long rx_bytes;
> > +	unsigned long rx_dropped;
> > +
> > +	unsigned long tx_pkts;
> > +	unsigned long err_pkts;
> > +	unsigned long tx_bytes;
> 
> why not per-queue stats? per-port stats are expensive

multi-queue is not supported in this implementation, but will be considered.

Regards
Qi
> 
> > +	uint16_t port_id;
> > +	uint16_t queue_idx;
> > +	int ring_size;
> > +	struct rte_ring *buf_ring;
> > +};


More information about the dev mailing list