2015-12-07 03:30, Liu, Jijiang: > Hi Thomas, > > > -----Original Message----- > > From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com] > > Sent: Monday, December 07, 2015 11:17 AM > > To: Liu, Jijiang > > Cc: dev at dpdk.org > > Subject: Re: [dpdk-dev] [PATCH] doc: announce ABI change for struct > > rte_eth_conf > > > > 2015-12-07 11:01, Jijiang Liu: > > > +* ABI changes are planned for struct rte_eth_conf in order to support > > > + tunneling packet configuration in unified tunneling API. The release > > > 2.2 > > does not contain these ABI > > > + changes, but release 2.3 will, and no backwards compatibility is > > > planned. > > > > Please, more details would be appreciated. > > We need to decide whether an ABI deprecation is the right choice. > > * ABI changes are planned for struct rte_eth_conf in order to support > tunneling packet configuration in unified tunneling APIs, which is the > rte_eth_dev_tunnel_configure > (uint8_t port_id, uint16_t rx_q, uint16_t tx_q, rte_eth_tunnel_conf * > tunnel_conf) API is planned to add. > and the 'tunnel_conf' shloud be stored in global 'rte_eth_conf'. > The release 2.2 does not contain these ABI change, but release 2.3 will, > and no backwards compatibility is planned. > > Is it enough clear?
No, I think we need an explanation in the commit message of what is the purpose of rte_eth_dev_tunnel_configure() and tunnel_conf. Ideally, an RFC patch would help.
